IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF ... · The evidence adduced at trial, and...

IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MASSACHUSETTS

In re: NEXIUM (ESOMEPRAZOLE) ANTITRUST LITIGATION

MDL No. 2409 Civil Action No. 1:12-md-02409-WGY

This Document Relates To: All Actions

PLAINTIFFS’ OPPOSITION TO DEFENDANTS’ RULE 50 MOTION

Case 1:12-md-02409-WGY Document 1190 Filed 11/10/14 of 57

i

TABLE OF CONTENTS

I. INTRODUCTION ...............................................................................................................1

II. PROCEDURAL BACKGROUND ......................................................................................6

III. LEGAL STANDARD ..........................................................................................................8

IV. ARGUMENT .......................................................................................................................9

A. Actavis describes a “large” payment as one that exceeds future litigation costs. .........................................................................................................9

B. The qualitative evidence shows that AstraZeneca’s payment to Teva was large. ......................................................................................................10

1. The indivisible deal that resolved both the Nexium and Prilosec litigations included a large payment from AstraZeneca to Teva. .................................................................................10

a. AstraZeneca and Teva were unable to settle Prilosec in 2006. ..........................................................................................10

b. In May 2008, AstraZeneca and Teva discuss settling both Nexium and Prilosec. .............................................................11

c. In July 2009, Teva suggested wrapping up the “prazole” litigations. ......................................................................13

d. Mr. Hester drafted and circulated the Nexium and Prilosec agreements together. ........................................................14

e. Mr. Hester hid that the Nexium and Prilosec settlements are a single enmeshed transaction that included a large payment. ..............................................................17

2. The jury is entitled to discredit testimony that simultaneously negotiating, drafting, and executing the Nexium and Prilosec agreements was merely convenient happenstance. .............................................................................................18

3. Teva abruptly quit the Nexium litigation, after having vigorously pursued it, in exchange for a large payment. ...........................20

a. In April 2008, Teva filed a declaratory judgment action to clear the way for Teva’s generic Nexium. ......................21

b. In 2007 and 2008, Teva rejected AstraZeneca’s May 27, 2014 generic Nexium entry date. .....................................21

c. In July 2009, Teva’s position in the Nexium case was strong. .....................................................................................22


ii

4. Teva resolved its material exposure in the Prilosec litigation for a non-material amount. .........................................................24

C. The quantitative evidence shows AstraZeneca’s payment to Teva exceeded AstraZeneca’s saved litigation costs. .....................................................25

1. Teva was exposed to substantial damages in the Prilosec case. ............................................................................................................27

a. Teva launched generic Prilosec at risk, in the face of courts deciding that other generics infringed AstraZeneca’s Prilosec patent. .......................................................27

b. In 2009, when AstraZeneca and Teva settled, the only question in the Prilosec litigation was precisely how much Teva would have to pay in damages. ...........................28

c. Teva’s position was that a reasonable royalty rate was somewhere between zero and 100%. ......................................29

d. AstraZeneca’s position was that it was entitled to a reasonable royalty rate plus enhanced damages, attorneys’ fees and costs, and prejudgment interest. ......................29

e. Teva considered the possibility of having to pay AstraZeneca more than a reasonable royalty rate devastating. ....................................................................................30

2. The jury can calculate the size of the payment in multiple ways. ..........................................................................................................32

a. Dr. McCool’s reasonable royalty rate: Teva owed AstraZeneca at least $42 million. ...................................................33

b. Teva’s license from Impax: As a floor, Teva would have paid AstraZeneca more for a risk-free license than it paid Impax to launch at risk. ...............................................36

c. Ms. Julie’s 2006 profit report: Teva’s Prilosec exposure ranged from $44 million to over $100 million. ...........................................................................................37

d. Teva’s reports to Ivax and AstraZeneca’s 80/20 deals: Teva’s Prilosec exposure ranged from over $30 million to about $70 million. ..................................................42

e. Dr. McGuire’s avoided litigation costs: AstraZeneca saved between $3 million - $5 million in litigation costs by settling the Nexium and Prilosec litigations. .........................................................................43

3. A quantitative comparison between the size of the payment and AstraZeneca’s Nexium sales is irrelevant and improper. ....................44


iii

V. CONCLUSION ..................................................................................................................46


iv

TABLE OF AUTHORITIES

Page(s)

Federal Cases

Alvord-Polk, Inc. v. F. Schumacher & Co., 37 F.3d 996 (3d Cir. 1994) .......................................20

Am. Graphics Inst., Inc. v. Darling, 2003 U.S. Dist. LEXIS 9790 (E.D. Pa. May 20, 2003) ..................................................................................................................................19

Amgen Inc. v. Hoechst Marion Roussel, Inc., 314 F.3d 1313 (Fed. Cir. 2003) .............................23

Astra Aktiebolag v. Andrx Pharm., Inc., 222 F. Supp. 2d 423 (S.D.N.Y.2002), aff’d by 84 Fed. Appx. 76 (Fed. Cir. 2003) .......................................................................10, 27

Brunswick Cmp. v. United States, 36 Fed. Cl. 204 (Ct. Fed. Cl. 1996) ........................................32

Copp v. Atwood, No. Civ.03–288–JD, 2005 WL 139180 (D.N.H. Jan. 24, 2005) ..........................8

Elliott v. Target Corp., 2013 WL 2153539 (D. Nev. May 16, 2013) ............................................38

Energy Transp. Grp., Inc. v. William Demant AIS, 697 F.3d 1342 (Fed. Cir. 2012) ....................32

FTC v. Actavis, Inc., 133 S.Ct. 2223 (2013) .......................................................................... passim

Hanson v. Alpine Valley Ski Area, Inc., 718 F.2d 1075 (Fed. Cir. 1983) ......................................36

Havens Steel Co. v. Randolph Engineering Co., 813 F.2d 186 (8th Cir. 1987) ............................40

Heyne v. Nick's Am. Pancake & Cafe, Inc., 2013 WL 6047553 (N.D. Ind. Nov. 15, 2013) ..................................................................................................................................38

Hodgens v. General Dynamics Corp., 144 F.3d 151 (1st Cir. 1998) ............................................20

In re SeaChange Int’l, Inc., No. 02-12116-DPW, 2004 U.S. Dist. LEXIS 1687 (D. Mass. Feb. 6, 2004) ..................................................................................................................24

In re Urethane Antitrust Litig., No. 04-1616-JWL, 2013 U.S. Dist. LEXIS 69784 (D. Kan. May 15, 2013) ...........................................................................................................20

Jastremski v. United States, 737 F.2d 666 (7th Cir. 1984) ............................................................38

Kruse, Inc. v. United States, 213 F. Supp. 2d 939 (N.D. Ind. 2002) .............................................38


v

Lighting Ballast Control, LLC v. Philips Elecs. N. Am. Corp., 814 F. Supp. 2d 665 (N.D. Tex. 2011) ......................................................................................................................32

Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301 (Fed. Cir. 2009) ..........................................32

Nassif v. U.S. for Use of Bayer & Mingolla Const. Co., 187 F.2d 794 (1st Cir. 1951) ..........................................................................................................................................8

Neel v. Mid-Atlantic of Fairfield, LLC, No. 10-cv-405, 2012 U.S. Dist. LEXIS 111939 (D. Md. Aug. 9, 2012).................................................................................................41

Parrish v. Nat'l Football League Players Inc., 2009 WL 88484 (N.D. Cal. Jan. 13, 2009) ........................................................................................................................................38

Putnam Res. v. Pateman, 958 F.2d 448 (1st Cir. 1992) .............................................................8, 20

R & B Transp., LLC v. U.S. Dept. of Labor, 618 F.3d 37 (1st Cir. 2010) .....................................20

Radio Steel & Mfg. Co. v. MTD Prods., Inc., 788 F.2d 1554 (Fed. Cir. 1986) .............................36

Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133 (2000) ................................................19

Santiago Hodge v. Parke Davis & Co., 909 F.2d 628 (1st Cir. 1990) ............................................6

Standard Mfg. Co. v. United States, 42 Fed. Cl. 748 (Ct. Fed. Cl. 1999) ......................................32

Sullivan v. Nat’l Football League, 34 F.3d 1091 (1st Cir. 1994) ....................................................8

TSC Indus., Inc. v. Northway, 426 U.S. 438 (1976) ......................................................................24

TVT Records v. Island Def Jam Music Group, 412 F.3d 82 (2d Cir. 2005) ..................................19

U.S. v. Bailey, 707 F.2d 19 (1st Cir. 1983) ......................................................................................8

United States v. Cintolo, 818 F.2d 980 (1st Cir. 1987) ..................................................................19

United States v. Cisneros, 448 F.2d 298 (9th Cir. 1971) ...............................................................19

United States v. Curtis, 635 F.3d 704 (5th Cir. 2011) ...................................................................20

United States v. Dentsply Int’l, Inc., 399 F.3d 181 (3d Cir. 2005) ................................................20

Van Orman v. Am. Ins. Co., 680 F.2d 301 (3d Cir. 1982) .............................................................19

Velazquez-Fernandez v. NCE Foods, Inc., 476 F.3d 6 (1st Cir. 2007) .........................................20

Whitserve, LLC v. Computer Packages, Inc., 694 F.3d 10 (Fed. Cir. 2013) ................................32


vi

Federal: Statutes, Rules, Regulations, Constitutional Provisions

17 C.F.R. § 229.103 .......................................................................................................................24

35 U.S.C. 284 .................................................................................................................................31

Federal Rule of Civil Procedure 50 ...................................................................................6, 7, 8, 46

Other Authorities

Aaron S. Edlin, C. Scott Hemphill, Herbert J. Hovenkamp, Carl Shapiro, Activating Actavis ......................................................................................................................9

Hovenkamp, Anticompetitive Settlement of Intellectual Property Disputes, 87 MINN. L. REV. 1719, 1758 (2003) ............................................................................................45

Hovenkamp, H., M. Janis, M. Lemley, & C. Leslie, IP and Antitrust (2d ed. Supp. 2013) (“IP & Antitrust”), § 15.3a ..............................................................................................9

Measure and Elements of Damages Recoverable for Attorney's Negligence In Preparing Or Conducting Litigation—Twentieth Century Cases, 90 A.L.R.4th 1033............................................................................................................................................8

Memorandum and Order on Summary Judgment, In re Nexium Antitrust Litig., 12-md-02409, ECF No. 977 (D. Mass. Sept. 4, 2014) .................................................... passim


1

I. INTRODUCTION

Whether a payment is “large” is driven by the underlying economics described in the

Supreme Court’s Actavis1 decision and can be assessed both qualitatively and quantitatively.

The evidence adduced at trial, and reasonable inferences therefrom, permit a jury to conclude

that AstraZeneca made a large payment to Teva under either rubric.

The Supreme Court emphasized that the “relevant anticompetitive harm” posed by a

reverse payment is the brand company’s paying its generic competitor to “prevent the risk of

competition.”2 A large payment induces the generic challenger to “quit its patent invalidity or

noninfringement claim,” “give up the patent fight,” and “stay away from the patentee’s market.”3

This Court expressed a similar analysis in its summary judgment ruling.4

The Supreme Court characterized a “large” payment as one that exceeds “a rough

approximation of the litigation expenses saved through the settlement.”5 Dr. McGuire reached

the same conclusion from an economic perspective: “If you see a payment from the brand that

exceeds its expected litigation costs, then there is strong economic evidence that the settlement

constitutes a delay in the sense of harming consumers.”6

1 FTC v. Actavis, Inc., 133 S.Ct. 2223 (2013). 2 Actavis, 133 S. Ct. at 2236. See also id. at 2230; and at 2234. The underlying rationale of Actavis is that

substance triumphs over form and that courts should avoid formulaic answers to substantive antitrust questions. 3 Id. at 2233. 4 See Memorandum and Order, ECF No. 977, at 116-17 (Sept. 4, 2014) (“SJ Order”) (rejecting Teva’s argument

that, if payment via the Prilosec settlement was $22.1 million, such a payment was “per se” lawful and stating that “The central question . . . is whether” the savings afforded to Teva by AstraZeneca in Prilosec “constituted a significant foregiveness of debt intended to induce Teva to delay its entry into the market for generic Nexium”) (emphasis added). See also id. at 54 (payment to Ranbaxy was “large and unjustified” if “Ranbaxy was induced to delay its generic launch in return”) (emphasis added).

5 133 S. Ct. at 2236. 6 11/7/14 Tr. 88:14-16.


2

Here, the jury can readily conclude from both qualitative and quantitative evidence that

AstraZeneca made such a payment to Teva.

Qualitative evidence. The evidence shows that the Nexium and Prilosec settlement

agreements were a single, indivisible, high-level deal that bolted resolution of the Nexium case

to a full release for Teva’s $100 million-plus exposure in the Prilosec case:

AstraZeneca and Teva signed the Nexium and Prilosec agreements on the same day after simultaneous negotiations among the same counsel.

Emails between AstraZeneca and Teva – involving the same AstraZeneca lawyers who negotiated the linked AstraZeneca-Ranbaxy Nexium settlement/side deals – always attached drafts of both agreements.

AstraZeneca and Teva were unable to settle the Prilosec litigation until it was tied to resolution of the Nexium litigation.

AstraZeneca’s outside counsel, Timothy Hester, drafted the Prilosec agreement on the same day that he said the parties reached a deal in principle on Nexium.

Teva rejected the May 2014 generic Nexium entry date twice, and did an about-face only after AstraZeneca offered it the Prilosec deal.

The negotiation and resolution of both litigations at once allows the jury to conclude that

AstraZeneca gave Teva a sweetheart Prilosec resolution in exchange for delayed entry in

Nexium. Two separate, stand-alone settlements would reflect7 a legitimate compromise of each

claim, with each party getting something and giving up something in each settlement. But

defendants’ insistence on resolving both litigations in a single deal permits the jury to conclude

that something else was going on: AstraZeneca got what it wanted in the Nexium piece of the

deal and gave up something in the Prilosec piece of the deal; Teva gave up something in the

Nexium piece of the deal and got what it wanted in the Prilosec piece of the deal. AstraZeneca

definitely got what it wanted in the Nexium piece: Mr. Hester told the jury that getting Teva to

7 Assuming nothing else amiss in each separate settlement.


3

accept the May 27, 2014 generic Nexium entry date meant “effectively we had won.”8 It follows

that Teva got something substantial in the Prilosec piece of the deal.

AstraZeneca and Teva’s deal fused the Nexium and Prilosec agreements, despite

AstraZeneca knowing that the bundled AstraZeneca-Ranbaxy agreements triggered FTC antitrust

scrutiny. The parties obviously needed to resolve both together as part of a single deal in order

to effectuate a large payment to Teva.9

That conclusion is confirmed by the very identity of the high level inside and outside

counsel who shepherded the joint deal: Did it really take the general counsel of AstraZenca (Mr.

Pott), the general counsel of Teva (Mr. Egosi) and the chairman of Covington & Burling (Mr.

Hester) to settle what the defendants say was the routine damages phase of a $9 million patent

case? Or did it take those high-level lawyers to approve, engineer, and try to disguise a payment

sufficiently large to induce Teva to quit the Nexium patent fight?

That AstraZeneca made such a payment is confirmed by the overwhelming evidence that

Teva abruptly reversed course and quit the Nexium patent fight:

Teva’s business model was to get an earlier entry date than the first ANDA filer, i.e., before May 27, 2014.

In 2007 and 2008, Teva rejected AstraZeneca’s offer of the May 27, 2014 entry date – the same offer Teva accepted after being paid off.

Teva did not suffer any adverse rulings or events in the Nexium infringement case between May 2008 and the time it quit.

AstraZeneca wanted to settle the Nexium case before the Markman hearing.

8 11/4/14 Tr. 59:11 (Hester). 9 If AstraZeneca and Teva had compromised on the Nexium entry date based on their respective views of the

underlying patent merits, as the FTC had long permitted, there would be no reason to resolve both cases in a single deal. The defendants’ need to resolve Nexium and Prilosec litigations at once arises because the Nexium settlement is not a compromise based on each party’s view of the merits, as Mr. Pott testified.


4

Though Teva claimed it was running short on time to break the bottleneck, Teva settled Nexium in principle in July 2009 – nearly five years before the May 2014 entry date.

This qualitative evidence, and reasonable inferences therefrom, permit the jury to conclude that

AstraZeneca’s payment to Teva was large. It exceeded any reasonable estimate of saved

litigation costs and was therefore sufficient to induce Teva to quit the Nexium fight that it had

been aggressively waging – precisely the anticompetitive harm that Actavis identified.

Quantitative evidence. Before addressing the quantitative evidence that the payment to

Teva exceeded AstraZeneca’s expected future litigation costs, a few points are worth noting:

Neither AstraZeneca nor Teva settled the Nexium case based on its assessment of the patent merits.

Neither AstraZeneca nor Teva settled the Prilosec case based on any apparent analysis of the damages in that case.

Neither AstraZeneca nor Teva intended the Prilosec settlement to merely reflect AstraZeneca’s saved litigation costs in the Nexium case.

Thus, while the plaintiffs can prove quantitatively that the payment exceeded expected

future litigation costs, the plain truth is that defendants do not even pretend that the payoff had

anything to do with saving those costs.

In any event, substantial evidence shows that the payment from AstraZeneca to Teva

exceeded AstraZeneca’s expected litigation costs:

AstraZeneca saved $3-4 million in litigation costs by settling the Nexium case,10 and a million dollars or less by settling the Prilosec litigation.11

Using Teva’s 2006 profit figures and the most conservative calculations, the jury could find that Teva was exposed to a judgment of well over $100 million; even if one discounts entirely AstraZeneca’s forceful efforts to collect treble damages, Teva’s exposure was well over $50 million.

10 11/7/14 Tr. 130:1-10 (McGuire). 11 11/7/14 Tr. 130:11-16 (McGuire)


5

Dr. McCool testified that a reasonable royalty would be at least 55% of Teva’s net generic Prilosec sales plus at least a $10 million up-front payment;12 As this Court has stated, “[y]ou fill in the blanks and arithmetic can then be done”13 to translate that into single damages of between $42 million and $57 million (not including enhanced damages).

AstraZeneca released Teva from its Prilosec liability for only $9 million.14

The evidence and reasonable inferences therefrom permit the jury to find that AstraZeneca paid

Teva far more than AstraZeneca’s saved litigation costs.15

Defendants’ error. Defendants argue that a payment’s size should be assessed by

comparing it to Nexium’s brand sales. This argument is contrary to Actavis. The alleged reverse

payment in Actavis was small if compared to the brand sales.16 Brand prices and profits are far

larger than those of generic companies – that is precisely why brands are able to pay off their

generic competitors to delay generic entry and why such payoffs trigger antitrust scrutiny.17

Actavis held that reverse payment cases such as this one present a quintessential fact

question: “Although the parties may have reasons to prefer settlements that include reverse

payments, the relevant antitrust question is: What are those reasons?”18 As the Supreme Court

then explained: “If the basic reason is a desire to maintain and to share patent-generated

monopoly profits, then, in the absence of some other justification, the antitrust laws are likely to

12 11/5/14 Tr. 124:9-11 (McCool); id. at 125:18-20. 13 11/5/14 Tr. 133:17-22 (McCool). 14 Exh. 46 at AZ-NX-MDL-00002540, Article 4. 15 While defendants contest the size of the reverse payment, that itself is a factual dispute that the jury must

decide. 16 See Actavis, 133 S. Ct. at 2229 (brand company’s payments totaled a few percentages of the brand sales

during the period of alleged delay). 17 For example, Teva’s forecaster, Ms. King, testified that Teva expects that generic prices with multiple generic

competitors will drop to 5% of the brand price. 10/29/14 Tr. 97:21-99:7 & Exh. 58. Indeed, prices can go even lower. See 10/31/14 Tr. 66:19-67:3 (generic prices can drop to 1% of brand with multiple generics) (Julie).

18 133 S. Ct. at 2237.


6

forbid the arrangement.”19 Here, the evidence and reasonable inferences therefrom permit the

jury to conclude that AstraZeneca and Teva entered into a single transaction to resolve both the

Prilosec and Nexium litigations in order to maintain and share AstraZeneca’s monopoly profits

(as articulated by the jury verdict form).

This Court denied the defendants’ motion for summary judgment regarding whether

AstraZeneca made a “large and unjustified”20 reverse payment to Teva.21 The evidence adduced

at trial is far stronger than the summary judgment record. The question “whether AstraZeneca

made a large payment to Teva” is factual, and remains one for the jury.22

II. PROCEDURAL BACKGROUND

The Court twice addressed the subject matter of the defendants’ motion under Federal

Rule of Civil Procedure 50 for judgment as a matter of law on whether AstraZeneca made a large

and unjustified payment to Teva.

At the September 30 hearing, the Court explained that the plaintiffs must put in evidence

of the “large and unjustified payment” to Teva first at trial, and that at the conclusion of the

plaintiffs’ evidence on this issue defendants may move for “summary judgment:”23

We agreed upon a six-week trial, a six-week trial we’re going to have, but you’re going to have to put in the Teva stuff first. I want all the stuff from the plaintiff’s point of view on substantial and unjustified payment arising out of that Teva . . . AstraZeneca agreements and how they were

19 Id. 20 Id.. 21 See generally SJ Order. 22 11/7/14 Tr. 59-63. See also Santiago Hodge v. Parke Davis & Co., 909 F.2d 628, 634 (1st Cir. 1990) (“The

standard of review for a motion for directed verdict is similar to that applied to a request for summary judgment”); id. (“It is well settled that upon the consideration of a motion for directed verdict, the evidence must be viewed in the light most favorable to the non-movant, giving him the benefit of every favorable inference that may be fairly drawn therefrom.”).

23 Plaintiffs respectfully maintain their objection to the requirement to “front load” the case and to the Court’s earlier summary judgment ruling regarding Ranbaxy causation. Plaintiffs also note that this opposition is being filed without benefit of time to review defendants’ brief.


7

implemented and I want that first. And when you’re done, I want to know it because the rule provides that when we have all the evidence on an issue -- and we’ll have all the plaintiff’s evidence on an issue . . . the defendants can move for summary judgment..24

At the October 15 charge conference, the Court further explained its thinking:

THE COURT: Look, this is how this is going to play out. You’ve given me no -- I’ll now amend Question 1 so it says “in a large and unjustified payment by AstraZeneca to Teva,” and I’ve already ordered that that’s the evidence I’m going to hear first. Once I’ve heard all that evidence, I am not bifurcating the trial we’ve already bifurcated.

MR. SOBOL: Sure. THE COURT: Once you’ve put in all your evidence on that point,

then you’re to notify the Court. That’s the pretrial order. I will expect, because they seem to be champing at the bit here, they will then move for a directed verdict. I will entertain argument on that motion and I will resolve it. Now, if I resolve it against the plaintiffs, I will be satisfied that I have created a sufficient record and I assume I will be satisfied with my ruling that the case is teed up for appeal. If I deny their motion for a directed verdict at that juncture, the case will continue to the point where the plaintiffs rest entirely. I do agree with you that I think it is up to the defense to come up with the procompetitive justifications.

MR. SOBOL: Sure. THE COURT: That’s true. And so when you rest, if you, um,

eluded directed verdict at that earlier stage in the trial, at least on that point and there would be others, causation and the like, you -- but on that point you probably will elude a directed verdict at the close of all your evidence. Then we’ll see what the defense comes up with as procompetitive justifications which, I’ve already said this, you not only will cross-examine, but you may then come up with, if it’s genuine, genuine rebuttal. Now that’s how I expect to manage the trial. But I’m going to leave the word “unjustified” in there.25

Given the Court’s statements about what is, and what is not, the subject of the

defendants’ rule 50 motion, plaintiffs are not addressing here – and have not finished presenting

24 9/30/14 Hearing Tr. at 5-6. 25 10/30/14 Hearing Tr. at 9-11 (emphasis added).


8

evidence at trial about – market power, antitrust impact, causation, damages, or other matters.

Nor have the plaintiffs tried to imagine and rebuff any-and-all potential justifications the

defendants may offer in their case in chief. A further discussion of defendants’ burden regarding

procompetitive justifications is included in Appendix A.

III. LEGAL STANDARD

Courts should grant directed verdicts only when the evidence “is such that reasonable

persons could reach but one conclusion.”26 Courts should not grant a directed verdict in a rule of

reason case even if the court is skeptical of an expert’s opinion.27 And the issue of whether a

payment is large is a question of fact for the jury.28

A comprehensive discussion of the relevant case law is attached as Appendix B.

26 Putnam Res. v. Pateman, 958 F.2d 448, 459 (1st Cir. 1992). Although this statement was made in the context

of a judgment notwithstanding the verdict (or “J.N.O.V.”), the Court makes it clear that the standards for directed verdict and “J.N.O.V.” are the same. Putnam Res., 958 F.2d at 459.

27 Sullivan v. Nat’l Football League, 34 F.3d 1091, 1105-06 (1st Cir. 1994). 28 11/7/14 Tr. 59:19-60:4 (The Court “the Supreme Court uses the word ‘large,’ and they don’t define it, you

know, they don’t, they say the other language which I read to the jury, and really the genius of our system, and I will tell them right now, ‘You know, folks, you’ve got to decide this as a matter of fact.’”). As another example, the issue of “fair value” and the value of a legal claim both are considered fact questions to be resolved by a jury. See Nassif v. U.S. for Use of Bayer & Mingolla Const. Co., 187 F.2d 794, 796 (1st Cir. 1951) (“The fair value … is a pure question of fact.”); U.S. v. Bailey, 707 F.2d 19, 21 (1st Cir. 1983) (“the fair market value of shares of stock is generally considered a question of fact that may not be disturbed unless it is clearly erroneous.”); John E. Theuman, Measure and Elements of Damages Recoverable for Attorney's Negligence In Preparing Or Conducting Litigation—Twentieth Century Cases, 90 A.L.R.4th 1033 (“Just what amount the client would have recovered in the prior action but for the attorney’s alleged negligence is generally a question for the factfinder in the malpractice trial”); Copp v. Atwood, No. Civ.03–288–JD, 2005 WL 139180, at *4 (D.N.H. Jan. 24, 2005) (granting summary judgment for plaintiff on question of attorney’s liability for malpractice, thus “leaving the issue of damages to be resolved either by the parties in settlement or, if necessary, by a jury.”).


9

IV. ARGUMENT

A. Actavis describes a “large” payment as one that exceeds future litigation costs.

The Supreme Court held that an antitrust defendant may be able to defend a payment by

showing that it did not exceed the patent holder’s reasonably expected litigation costs, a

quantitative measure.29 This is consistent with academic literature:

The reason the patentee is willing to make a payment that exceeds its litigation costs is precisely because the settlement will permit it to exclude competition from the market, whereas if it went to trial there is a chance that the patent would be held invalid or not infringed and the market would become competitive. On expectation, the patentee is paying for an advantage it could not get if it went to trial.30

Likewise, Activating Actavis proposes jury instructions emphasizing the significance of

avoided litigation costs: “In assessing whether this payment is unreasonably large, you may

consider whether the payment is no greater than the patent holder’s anticipated litigation costs

that are avoided through settlement.”31 As Dr. McGuire explained, no brand manufacturer will

pay more to a generic than the brand would expect to pay in future litigation costs unless the

brand concludes that it can get a better deal by paying its competitor – which means obtaining

more protection from competition than it expects its patents alone would afford if fully

litigated.32 That size payment is plainly “large.”33

29 Actavis, 133 S. Ct. at 2236-37. 30 Hovenkamp, H., M. Janis, M. Lemley, & C. Leslie, IP and Antitrust (2d ed. Supp. 2013) (“IP & Antitrust”),

§15.3a (emphasis added). 31 Aaron S. Edlin, C. Scott Hemphill, Herbert J. Hovenkamp, Carl Shapiro, Activating Actavis, 26 ANTITRUST

16, 21 (Fall 2013). 32 11/7/14 Tr. 86:12-20 (“[W]hat I can say, if you see a payment, that means that date has been delayed. The

brand had that going in. They put in some money, they made the payment. The way to understand that from an economic point of view is they’re getting something out of it. They’re getting longer time to sell without generic competition.”) (Dr. McGuire).

33 Dr. McGuire explained to the jury that a payment is “large” in economic terms if it exceeds the brand company’s avoided litigation costs. See 11/7/14 Tr. 89:25-91:2.


10

B. The qualitative evidence shows that AstraZeneca’s payment to Teva was large.

1. The indivisible deal that resolved both the Nexium and Prilosec litigations included a large payment from AstraZeneca to Teva.

The evidence, and reasonable inferences therefrom, permit the jury to conclude that the

Prilosec and Nexium settlements were a single, indivisible, high-level deal that bolted resolution

of the Nexium case to a full release for Teva’s $100 million-plus exposure in the Prilosec case.

The Prilosec litigation long pre-dated the Nexium litigation. Teva launched a generic

version of Prilosec – “one of the most widely prescribed medicines in history”34 – in September

2004 “at risk.” AstraZeneca sued Teva for infringement in January 2005.35 But the Prilosec

litigation started long before, in the early 2000s; AstraZeneca sued Teva’s partner Impax in

2000.36 AstraZeneca won on patent validity and infringement against multiple other companies

in 2002, a decision affirmed on appeal by the Federal Circuit in 2003, a year before Teva

launched at risk.37

As to Nexium, Teva filed its Paragraph IV certification as the second filer, after Ranbaxy,

in November 2005. AstraZeneca sued Teva for infringement in March 2006.38

a. AstraZeneca and Teva were unable to settle Prilosec in 2006.

Teva and AstraZeneca tried settling Prilosec in 2006.39 On April 18, 2006, two weeks

after a bench trial in the Prilosec patent case began,40 Teva sent AstraZeneca financial

34 Exh. 52 at 390. 35 Exh. 1, ¶¶ 101-02. 36 Exh. 1, ¶100. 37 Exh. 52 at 390 n.2. See also Astra Aktiebolag v. Andrx Pharm., Inc., 222 F. Supp. 2d 423 (S.D.N.Y.2002),

aff’d,84 Fed. Appx. 76 (Fed. Cir. 2003) (finding that Genpharm, Cheminor, and Andrx infringed AstraZeneca’s Prilosec patents).

38 Exh. 1, ¶¶ 78-79 (the Paragraph IV certification was actually filed by Ivax, which was acquired by Teva in January 2006, see id.); 10/27/14 Tr. 140:23-141:9 (Pott). See also 10/28/14 Tr. 23:19-24 (“Teva, at some point, launched in 2004, I believe, and we sued them in 2005.”) (Pott).

39 Exh. 48 (email from Staci Julie of Teva to Marcus Heifetz of AstraZeneca dated April 18, 2006).


11

information to facilitate a settlement discussion.41 This 2006 data showed Teva profits nearly

twice as large as the defendants now contend they were.42 AstraZeneca’s General Counsel, Mr.

Pott, however, testified “I don’t recall” when asked to explain AstraZeneca’s view in 2009 of the

“fair value” of a Prilosec settlement with Teva;43 while he said AstraZeneca had formed a view

of Teva’s generic Prilosec profits back then, he claimed he could not recall that view either.44

The evidence, and reasonable inferences therefrom, permit the jury to conclude that AstraZeneca

did not perform any Prilosec damages analysis, or use any data at all during its negotiations with

Teva.

These early settlement talks between AstraZeneca and Teva went nowhere until Nexium

settlement talks began.45

b. In May 2008, AstraZeneca and Teva discuss settling both Nexium and Prilosec.

In May 2008, the day after Teva filed its Nexium Declaratory Judgment action against

AstraZeneca, Ms. Julie, her boss David Stark, Richard Egosi, and another Teva attorney met in

Philadelphia with AstraZeneca’s General Counsel Mr. Pott and other AstraZeneca lawyers.46

David Stark is currently acting General Counsel of Teva’s parent company and in 2008-09 was

40 Exh. 52 at 390 (“The case was tried to the Court sitting without a jury for 42 trial days, stating April 3, 2006

and ending June 14, 2006.”); 10/27/14 Tr. 160 (Pott). 41 Exh. 48 (email from Staci Julie of Teva to Marcus Heifetz of AstraZeneca dated April 18, 2006). 42 Id. 43 10/28/14 Tr. 49:3-10 (‘Yeah, I don’t recall what the internal view was.”). 44 10/28/14 Tr. 49:12-16 (“I think we had a range that we understood their profits so that I could negotiate that

in the context of the information I had. Q. So what was the range? A. I don’t recall the range.”). 45 10/28/14 Tr. 58:18-61:3 (Pott agreed that the Prilosec settlement discussions started when Teva and

AstraZeneca were discussing the Nexium litigation, but could not explain why). In early 2008 or late 2007, Teva (through its in-house lawyer Ms. Julie) contacted AstraZeneca’s general counsel (Mr. Pott) about settling both the Prilosec and Nexium lawsuits. 10/27/14 Tr. 143:23-144:5 (Pott); 10/30/14 Tr. 31:22-32:9 (Julie), Exh. 65 (Ms. Julie wrote to Mr. Pott that she, Mr. Egosi, Mr. Stark, and Mr. Yaari planned to attend the May 1, 2008 meeting with AstraZeneca); 10/30/14 Tr. 24:22-25:3 (Ms. Julie testified that she reached out to Mr. Pott at the end of 2007).

46 10/30/14 Tr. 38:2-10 (Julie).


12

the General Counsel of Teva North America.47 In 2009-2010, Mr. Egosi was the Chief Legal

Officer of Teva’s parent company.48 At this May 2008 meeting in Philadelphia about settling the

Nexium litigation, Teva and AstraZeneca also discussed the Prilosec settlement.49

The jury could conclude that all these high-level lawyers – later including Mr. Tim

Hester, an antitrust (not patent) lawyer and chairman of the D.C. law firm Covington & Burling

LLP50 – were not needed just to resolve what defendants contend was a routine damages case

worth $9 million.

The Prilosec litigation had progressed very poorly for Teva (and very well for

AstraZeneca) since their fruitless settlement talks in 2006. In May 2007, the presiding court

ruled that AstraZeneca’s Prilosec patents were valid and infringed by Teva’s partner Impax,51

forcing Teva to discontinue its sales of generic Prilosec. The Federal Circuit affirmed the district

court’s decision in August 2008.52

Just as AstraZeneca was prepared to pursue the Prilosec patent case against Teva, Teva

was fully prepared to press the Nexium patent case against AstraZeneca.53 At the May 2008

meeting, AstraZeneca offered Teva the May 27, 2014 entry date, but Teva rejected it.54 Mr. Pott

confirmed that Teva’s position in May 2008 was that it was “planning to come to the market

47 10/30/14 Tr. 9:6-10 (Julie). 48 10/30/14 Tr. 10:12-15 (Julie). 49 10/27/14 Tr. 144:10-145:1, 162:2-8 (Pott); 10/30/14 Tr. 38:13-39:1 (Julie). 50 10/31/14 Tr. 161:10-17 (Hester). 51 Exh. 52; 10/31/14 Tr. 34:17-35:6 (Julie). 52 Exh. 53. As Mr. Pott testified, as of May 2008, “we had won the underlying case against Teva.” 10/28/14

Tr. 40:4-42:10. See also 10/30/14 Tr. 90:6-15 (Julie); 10/31/14 Tr. 36:1-2 (Julie). 53 10/30/14 Tr. 40:14-19 (Julie). 54 10/28/14 Tr. 64:24-65:3 (AstraZeneca offered Teva the May 2014 entry date in May of 2008 but Teva

rejected it) (Pott); 10/30/14 Tr. 40:14-20, 42:23-43:17 (Julie); 11/4/14 Tr. 111:6-10 (Hester); 10/30/14 Tr. 42:17-18 (Julie).


13

[with its generic Nexium] prior to the expiration of our [Nexium] patents.”55 Teva told

AstraZeneca that “they were going to keep moving” the Nexium litigation forward, and “in fact,

did keep moving it.”56 Mr. Pott understood that Teva’s goal was still to get an earlier entry date

than Ranbaxy.57

c. In July 2009, Teva suggested wrapping up the “prazole” litigations.

Just a little over a year later, with a Markman hearing in the Nexium case approaching,

Teva’s general counsel (Egosi) contacted his counterpart at AstraZeneca (Pott) in July 2009

about “wrapping up our prazole litigations” (meaning the Prilosec and Nexium cases).58

AstraZeneca’s counsel responded by telling Teva that AstraZeneca “want[s] to postpone the

upcoming [Markman] hearing so we have some time ‘to discuss.’”59

By July 2009 in the Prilosec case, AstraZeneca had won both at the district court level

and in the court of appeals.60 Teva reported in its 2004 20-F that “Were AstraZeneca ultimately

to be successful in its allegation of patent infringement, Teva could be required to pay damages

related to a portion of the sales of Impax’s omeprazole capsules and be enjoined from selling that

55 10/27/14 Tr. 146:14-18 (Pott). See also id. at 147:12-14 (“yeah, they were saying they were going to litigate

the case and try to invalidate or show noninfringement of the patents.”) (Pott); 10/28/14 Tr. 64:20-65:14 (Pott). See also 10/28/14 Tr. 65:9-67:6 (Pott).

56 10/27/14 Tr. 146:9-10 (Pott). 57 10/28/14 Tr. 65:14-22 (Pott). 58 10/27/14 Tr. 149:15-150:24 (Pott) & Exh. 38 (email from Egosi to Pott dated July 2, 2009) (emphasis added).

Mr. Egosi wrote: “Anything new from your end on wrapping up our prazole litigations?” Exh. 38. Mr. Pott testified that by saying “prazole,” Mr. Egosi was combining omeprazole (generic Prilosec) with esomeprazole (generic Nexium) into his own shorthand for both cases. See also 10/30/14 Tr. 86:5-13 and Exh. 75 (in an August 4, 2009 email to Mr. Pott and Mr. Heifetz (and others), Mr. Egosi writes “[w]e can discuss both esomep and omep.”). Ms. Julie explained that this referred to generic Nexium and generic Prilosec. 10/30/14 Tr. 87:9-20 & Exh. 76 (Teva General Counsel David Stark wrote to internal counsel at Teva “Can you guys join this call tomorrow to prep the settlement call with AZ re: esomep and omep…”); 10/30/14 Tr. 38:13-19 (the agenda for a May 1, 2008 meeting between AstraZeneca and Teva’s general counsels included both Nexium and Prilosec); 10/30/14 Tr. 38:20-39:1 (both Nexium and Prilosec were discussed at a May 1, 2008 meeting).

59 Exh. 39 at Teva-ESO-024316 (email from Egosi to Pott dated Aug. 3, 2009). 60 10/28/14 Tr. 41:17-42:10 (Pott). See Exh. 53 (decision of the Federal Circuit issued Aug. 20, 2008).


14

product.”61 While Teva may have had an argument that it was not bound by the final judgment

of infringement against Impax, it was only a technical argument. Teva sold the very Impax

generic Prilosec product that was found to infringe.62 As a result, whether Teva was technically

bound, or the judgment was applied to it as a matter of stare decisis, the only real issue was the

amount of damages Teva owed AstraZeneca.63 Mr. Pott agreed that AstraZeneca had a “strong”

case against Teva on Prilosec,64 and AstraZeneca’s position was that Teva was liable for an

amount “not less than a reasonable royalty and enhanced damages . . . because Teva willfully

infringed”65 plus attorneys’ fees.66

In contrast, Teva’s position in the Nexium case in July 2009 was strong (see infra section

IV.B.3.c).

d. Mr. Hester drafted and circulated the Nexium and Prilosec agreements together.

After initially asserting that the parties first reached an agreement in principle in Nexium

(in July 2009) and later discussed a Prilosec resolution,67 Mr. Hester conceded that his privilege

log showed that on July 30, 2009, he worked on both a draft Nexium agreement68 and a draft

Prilosec agreement.69

61 Exh. 110 at F-32 (emphasis added). 62 Exh., 1, ¶¶ 101, 103; 10/30/14 Tr. 70:15-20 (Julie). 63 10/28/14 Tr. 41:17-42:10 (Pott). See Exh. 53 (decision of the Federal Circuit issued Aug. 20, 2008). 64 10/28/14 Tr. 55:20-22 (“Yes, we had a strong -- we had an underlying finding of infringement and we had

removed them from the marketplace. It was strong, yes.”). 65 Exh. 44 at AZ-NX-MDL01016965 (emphasis added); 10/28/14 Tr. 31:13-32:9 (Pott). 66 Exh. 43 at AZ-NX-MDL-01016203-04. AstraZeneca’s position was that “Teva . . . is liable to the same

extent as Impax.” Id. at AZ-NX-MDL-01016197. 67 11/4/14 Tr. 92:23-94:1 (Hester). 68 11/4/14 Tr. 96:25-97:20 (Hester); id. at 98:13-17. See also Exh. 98 (privilege log reflecting Hester drafted

Prilosec agreement on July 30, 2009). 69 11/4/14 Tr. 97:21-98:12 (Hester). See also Exh. 98 (privilege log reflecting Hester drafted Nexium

agreement on July 30, 2009).


15

Mr. Pott and Mr. Egosi spoke on August 20, 2009 regarding both the Prilosec and

Nexium cases.70 Less than a week after that call, Mr. Hester circulated draft settlement

agreements for both lawsuits71 providing that (as to Nexium) Teva agreed to the same May 27,

2014 entry date72 as Ranbaxy – the very date Teva rejected in 200773 and 2008.74 Regarding

Prilosec, AstraZeneca agreed to accept a $9 million payment75 to release Teva from liability that

the jury is entitled to value at more than $100 million (discussed further below). The $9 million

was just over half of the amount of attorneys’ fees AstraZeneca asserts it spent litigating the

Prilosec matter ($15 million).76

A jury could conclude that because Teva accepted the same date that it had twice rejected

earlier, Teva must be getting something else in order to settle.

The “prazole” settlement discussions were inextricably intertwined and culminated in a

single transaction. Between August 2009 and January 6, 2010, the day the deal was completed,77

every time a draft of the Nexium settlement was circulated, it was accompanied by a draft of the

70 Exh. 40; 10/28/14 Tr. 57:12-58:17 & Exhs. 38-39. 71 10/28/14 Tr. 61:6-62:7 (Pott) & Exh. 46. 72 Exh. 46 at AZ-NX-MDL-00002517, Article 5.2. 73 11/4/14 Tr. 109:11-21 (“Q. Mr. Hester, are you aware that Mr. Pott offered the May 27, 2014, entry date to

Teva back in 2008? A. I thought it was discussed earlier before Judge Pisano even in 2007. I think he may have -- he may have mentioned that date in that initial conference before Judge Pisano to discuss settling the cases. Q. For Teva? A. But Teva didn’t agree. Q. Fair enough. A. Because they were going ahead to try and litigate the cases.”) (Hester).

74 10/28/14 Tr. 64:24-65:3 (AstraZeneca offered Teva the May 2014 entry date in May of 2008 but Teva rejected it) (Pott); 10/30/14 Tr. 40:14-20, 42:23-43:19 (Julie); 11/4/14 Tr. 111:6-10 (Hester).

75 Exh. 46 at AZ-NX-MDL-00002540, Article 4. 76 Mr. Pott agreed with Terri Bowman’s testimony that AstraZeneca spent about $15 million litigating its

Prilosec claims against Teva. 10/28/14 Tr. 72:18-22; 10/29/14 Tr. 131:13-17. See Exh. 107 (Declaration of Terri L. Bowman).

77 Exh. 11 (AstraZeneca-Teva Nexium agreement); Exh. 12 (AstraZeneca-Teva Prilosec agreement); 10/30/14 Tr. 99:24-100:1 (Julie); 10/27/14 Tr. 140:23-141:9 (Pott); 10/28/14 Tr. 39:3-40:10; 10/31/14 Tr. 55:11-13 (Julie).


16

Prilosec settlement.78 The two negotiations and settlements were fused together into a single

deal.79 The two agreements were negotiated together by the same lawyers, drafted concurrently,

finalized at the same time,80 and executed together on the same day.81 Mr. Pott acknowledged

that it was “[n]o coincidence” that the settlements were signed the same day.82

It is also no coincidence that AstraZeneca used the same cast of attorneys to settle with

Teva as it did with Ranbaxy. The jury can conclude that the AstraZeneca and Ranbaxy

settlement agreement and side deals were part of a single indivisible transaction.83 The jury may

properly conclude that the same AstraZeneca players used the same phony-separate-deals ploy to

make a large payment to Teva.

78 See, e.g., Exh. 77 (Compilation of 20 cover emails for draft settlement agreements for Prilosec and Nexium

dated August 2009 through January 2010); & 10/30/14 Tr. 96:14-99:23 (Julie); 11/4/14 Tr. 74:9-13 (Hester); Exh. 47.

79 Exh. 77; 10/30/14 Tr. 96:14-99:23 (Julie); 10/30/14 Tr. 8:4-9 (Julie); Exh. 39; 10/27/14 Tr. 151:1-153:22 (Pott); 10/28/14 Tr. 58:15-17 (Pott). Even in denying the relationship between the Prilosec and Nexium litigations, Mr. Hester testified that the two were settled “together”: “That’s right. And they weren’t joined even then. The parties, my understanding was the parties were willing to settle each of them separately, but it made sense to look at whether they could settle them together. Because there were these two disputes, made sense to look at settling two disputes.” 11/4/14 Tr. 93:21-94:1.

80 See, e.g., 10/30/14 Tr. 99:19-100:10 (Julie); Exh. 77. 81 10/30/14 Tr. 99:24-100:1. 82 10/28/14 Tr. 75:2-5. Mr. Pott acknowledged that the signatures on the two settlements were obtained

remotely (as commonly occurs) and the individuals were not all physically in the same conference room. 10/28/14 Tr. 76:3-15. Thus, the jury can easily find that there was no “efficiency” reason for the simultaneous signing.

83 AstraZeneca admits that Ranbaxy was willing to delay launching its generic until 2012 in return for the “no authorized generic” promise, but AstraZeneca also agreed to a series of lucrative side deals with Ranbaxy, and Ranbaxy agreed to delay its generic launch even further until May 27, 2014. 10/27/14 Tr. 77-87, 95, 98:21-99:14; 118:16-122:18; 127:12-129:25 (Pott); Exhs. 31 (generic Plendil distribution agreement); 32 (generic Prilosec distribution agreement); 33 (Nexium API supply agreement); 34 (tolling agreement for making finished Nexium); 35 (bailment agreement). Ranbaxy made expressly clear that the settlement and side deals were a single transaction and that it would not sign the settlement without the side deals. See, e.g., 10/27/14 Tr. 106:17-108:17 (Pott) (discussing Exh. 27) (email exchange in which Ranbaxy explains that the API side deal “would be a required component to signing off on the Settlement Agreement and the two can be done in parallel.”).


17

e. Mr. Hester hid that the Nexium and Prilosec settlements are a single enmeshed transaction that included a large payment.

Tellingly, the defendants tried to hide that Nexium and Prilosec were resolved in a single

deal. On January 4, 2010, Mr. Hester84 sent Teva’s counsel a redline of Teva’s draft press

release announcing the Prilosec and Nexium settlements.85 The draft press release that Teva

prepared stated: “Teva Pharmaceutical Industries Ltd. (Nasdaq: TEVA) announced today that it

has entered into a [singular] definitive agreement [singular] with AstraZeneca to settle patent

litigation regarding Teva’s U.S. generic versions of AstraZeneca’s Prilosec® (omeprazole) and

Nexium® (esomeprazole), including all claims for patent infringement and damages.”86 Mr.

Hester proposed changing “a definitive agreement” to “two definitive agreements.”87 He also

suggested that Teva not disclose that it was paying AstraZeneca only $9 million.88

Mr. Hester’s experience representing AstraZeneca and Mr. Pott in then-ongoing FTC

proceedings involving the payments embodied in the earlier AstraZeneca-Ranbaxy agreements

undoubtedly informed his caution.89 The FTC’s investigation of the linked AstraZeneca-

84 11/5/14 Tr. 57:14-20 (Hester). 85 Exh. 77, FQT, AZ-NX-MDL-00024753-758. 86 Exh. 77, FQT, AZ-NX-MDL-00024753-758, at 756 (emphasis added). 87 Id.(emphasis added). Mr. Hester testified unconvincingly about this change that: “A. I mean they drafted it --

I just thought my change made it more accurate and clearer. It was really just a small drafting point.” 11/4/14 Tr. 80:16-18.

88 Exh. 77, FQT, AZ-NX-MDL-00024753-758, at 756; 11/4/14 Tr. 81:3-10 (Hester). With respect to this edit, Mr. Hester claimed: “we just thought there’s no reason to include dollar amounts in a press release like this, it's just an announcement of resolving two pieces of patent litigation, period, end of story.” 11/4/14 Tr. 81:14-17.

89 As Mr. Hester testified, he was brought in because he expected the Teva transaction could be subject to antitrust scrutiny. 11/5/14 Tr. 57:14-20 (“Q. When you were negotiating these transactions, Mr. Hester, did you have it somewhere in your mind that these transactions at some point would be subject to antitrust scrutiny? Yes or no. A. I was brought in as an antitrust lawyer and, yes, I was brought in to be considering the issues from an antitrust perspective, yes.”).


18

Ranbaxy agreements began in July 3, 2008; Mr. Pott gave a deposition on July 23, 2008 (and

was defended by Mr. Hester).90

This evidence and reasonable inferences therefrom permit the jury to find that efforts to

cover up the “package” nature of the “prazole” deals show that the Prilosec settlement disguised

a reverse payment. Why else would Teva and AstraZeneca have worked so hard to conceal and

deny the relationship between the agreements? The testimony from both AstraZeneca and Teva

witnesses is that the settlements were negotiated in tandem from start to finish and that Teva

accepted the May 2014 generic Nexium entry date only after AstraZeneca relented despite its

“strong” Prilosec position and agreed to settle the case for what amounts to just a portion of its

out-of-pocket attorneys’ fees and effectively zero damages.91

The evidence and reasonable inferences therefrom permit the jury to find that the

payment was “large” because the defendants combined the settlements into a single deal even in

the shadow of antitrust liability, and enlisted the assistance of each company’s general counsel

and the now Chairman of Covington & Burling.

2. The jury is entitled to discredit testimony that simultaneously negotiating, drafting, and executing the Nexium and Prilosec agreements was merely convenient happenstance.

Against this overwhelming evidence, the jury could easily reject, and need not credit,

AstraZeneca’s and Teva’s feeble assertions that the negotiations of the settlements were

completely independent of each other. Whether the Prilosec settlement was in fact “separate”

from the Nexium settlement is a classic jury question: “[W]hether multiple writings should be

90 Exh. 95, AstraZeneca’s First Civil Investigative Demand Responses dated May 29, 2009 (referencing date of

July 3, 2008 Civil Investigative Demand); Exh. 99, Excerpts from transcript of investigational hearing, Jeffrey A. Pott, FTC, dated July 23, 2008.

91 Subtracting the $9 million Teva paid from the $15 million in fees expended by AstraZeneca results in a loss to AstraZeneca of $6 million, and therefore effectively no recovery whatsoever, in a patent case it had won.


19

construed as one agreement depends upon the intent of the parties . . . an issue which is typically

a question of fact for the jury.”92

“A trier of fact is not compelled to accept and believe the self-serving stories of vitally

interested defendants. Their evidence may not only be disbelieved, but from the totality of the

circumstances, including the manner in which they testify, a contrary conclusion may be

properly drawn.”93 This is consistent with the “general principle of evidence law” that “the

factfinder is entitled to consider a party’s dishonesty about a material fact as ‘affirmative

evidence of guilt.’”94 Thus, particularly where other corroborative evidence exists, courts

frequently conclude that the falsity of a defendant’s testimony can be persuasive circumstantial

92 TVT Records v. Island Def Jam Music Group, 412 F.3d 82, 89 (2d Cir. 2005) (citations omitted); Van Orman

v. Am. Ins. Co., 680 F.2d 301, 306 (3d Cir. 1982) (“two or more writings may constitute a single contract even though they do not refer to each other. Whether two writings are to be construed as a single contract . . . depends on the intent of the parties”) (citations omitted); Am. Graphics Inst., Inc. v. Darling, 2003 U.S. Dist. LEXIS 9790, at *22-23 (E.D. Pa. May 20, 2003) (“The rule applies even where the parties are not the same, if the several instruments were known to all the parties and were delivered at the same time to accomplish an agreed purpose.”) (citing 17A Am. Jur. 2d Contracts § 388).

93 United States v. Cisneros, 448 F.2d 298, 305-06 (9th Cir. 1971) (citing Dyer v. MacDougall, 201 F.2d 265, 268 (2d Cir. 1952)). See also United States v. Cintolo, 818 F.2d 980, 989 (1st Cir. 1987) (citing Cisneros and finding a “jury was reasonably entitled to disbelieve [the defendant’s] testimony regarding his motives and to credit the (entirely plausible) contrary interpretation urged by the government”).

94 Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 147 (2000) (quoting Wright v. West, 505 U.S. 277, 296 (1992)).


20

evidence of liability or guilt.95 Acts of concealment may likewise provide circumstantial

evidence of the existence of a conspiracy.96

If the jury credits the plaintiffs’ evidence over the defendants’ (as it is free to do),97 the

jury may further find that Teva’s and AstraZeneca’s attempts to disguise the payment, and deny

that the forgiveness of Teva’s Prilosec litigation was a payment to settle the Nexium litigation, is

itself evidence that the payoff was large.98

3. Teva abruptly quit the Nexium litigation, after having vigorously pursued it, in exchange for a large payment.

Teva’s abrupt reversal of course in quitting the Nexium litigation – after having

vigorously pursued it for years – allows the jury to conclude that Teva received a large payment

in exchange for quitting.

95 See, e.g., R & B Transp., LLC v. U.S. Dept. of Labor, 618 F.3d 37, 47 (1st Cir. 2010) (affirming

administrative law judge’s conclusion that the employer’s lies justified a finding that its proffered reason for an adverse employment action was pretext for an unlawful termination); Velazquez-Fernandez v. NCE Foods, Inc., 476 F.3d 6, 13 (1st Cir. 2007) (noting “the Supreme Court has held that proof that the employer’s explanation is ‘unworthy of credence’ is one form of ‘circumstantial evidence that is probative of intentional discrimination’” (quoting Reeves, 530 U.S. at 147) and indicating that it would be proper to disbelieve explanations that are “inaccurate, unbelievable, idiosyncratic, or misleading”); Hodgens v. Gen. Dynamics Corp., 144 F.3d 151, 168 (1st Cir. 1998) (evaluating summary judgment ruling and stating that a non-moving employment discrimination plaintiff may demonstrate pretext by pointing out “‘such weaknesses, implausibilities, inconsistencies, incoherencies, or contradictions in the employer’s proffered legitimate reasons for its action that a reasonable factfinder could rationally find them unworthy of credence and [with or without additional evidence and inferences properly drawn therefrom] infer that the employer did not act for the asserted non-discriminatory reasons.’” (modification in original)).

96 See, e.g., United States v. Curtis, 635 F.3d 704, 717 (5th Cir. 2011) (“‘[A]cts of concealment done in furtherance of the main criminal objectives of the conspiracy’ are circumstantial evidence of the conspiracy’s existence” (citing, inter alia, United States v. Evans, 572 F.2d 455, 468 (5th Cir. 1978) (“explaining that because concealment is one of ‘the hallmarks of a conspiracy, . . . the objective and observable acts of the conspirators are relevant and competent circumstantial evidence from which the jury may draw the inference of the existence of the agreement or common purpose’”)); In re Urethane Antitrust Litig., No. 04-1616-JWL, 2013 U.S. Dist. LEXIS 69784, at *61-62 (D. Kan. May 15, 2013) (citing Curtis and finding relevant evidence that could show an attempt by the defendant to cover up its illegal activities).

97 Putnam Res., 958 F.2d at 459. 98 See generally Alvord-Polk, Inc. v. F. Schumacher & Co., 37 F.3d 996, 1012-13 (3d Cir. 1994) (noting

evidence that defendant was “attempting to disguise the true reason for its actions”); United States v. Dentsply Int’l, Inc., 399 F.3d 181, 196 (3d Cir. 2005) (“Dentsply’s asserted justifications for its exclusionary policies are inconsistent with its announced reason for the exclusionary policies, its conduct enforcing the policy, its rival suppliers’ actions, and dealers’ behavior in the marketplace”).


21

a. In April 2008, Teva filed a declaratory judgment action to clear the way for Teva’s generic Nexium.

AstraZeneca and Ranbaxy announced their Nexium deal in April 2008, prompting Teva

to take the aggressive step of filing a declaratory judgment action of its own against AstraZeneca

in order “to get judicial resolution as to all of the orange book patents” to clear the way for

Teva’s generic Nexium product to get to market.99 Teva’s in-house counsel, Staci Julie, testified

that if Teva did not act, and if Ranbaxy continued to hold first-to-file status, Teva would be

bottlenecked behind Ranbaxy.100

AstraZeneca’s payment to Teva must be evaluated in this context: AstraZeneca had

already paid off Ranbaxy, the first ANDA filer. It took less for AstraZeneca to buy off Teva, the

second filer, and induce it to join the conspiracy to delay generic competition.

b. In 2007 and 2008, Teva rejected AstraZeneca’s May 27, 2014 generic Nexium entry date.

Teva rejected AstraZeneca’s offer of the May 27, 2014 entry date in 2007 and again in

2008. Mr. Pott confirmed that Teva’s position in May 2008 was that it was “planning to come to

the market [with its generic Nexium] prior to the expiration of our [Nexium] patents.”101 Teva

told AstraZeneca that “they were going to keep moving” the Nexium litigation forward, and “in

fact, did keep moving it.”102

99 10/30/14 Tr. 28:15-18, 30:9-11, 33:14-17 (Julie); Exh. 66. 100 10/30/14 Tr. 41:6-15 (Julie). 101 10/27/14 Tr. 146:14-18 (Pott). See also id. at 147:12-14 (“yeah, they were saying they were going to litigate

the case and try to invalidate or show noninfringement of the patents.”) (Pott); 10/28/14 Tr. 64:20-65:14 (Pott); 10/28/14 Tr. 65:9-67:6 (Pott); 10/30/14 Tr. 38:20-39:1 (Julie).

102 10/27/14 Tr. 146:9-10 (Pott).


22

Litigating cases to get earlier entry dates is “[Teva’s] business model as a generic drug

maker.”103 Teva’s goal was to get an earlier entry date than Ranbaxy.104

Ms. Julie testified that the position that Teva takes in all cases in which it is the second-

filer (as in Nexium) is that Teva wants a “better” date than the first filer: “That was what we

were trying for. That’s why we litigate.”105 And this is what Teva told AstraZeneca: Teva

intended to break the bottleneck and get a better entry date.106

c. In July 2009, Teva’s position in the Nexium case was strong.

Teva made significant progress in the Nexium patent case. AstraZeneca wanted to

postpone the Markman hearing to try to settle.107 Ms. Julie tried to assert that Teva’s position

had been weakened by a stay of its declaratory judgment action, which would have slowed

Teva’s ability to trigger Ranbaxy’s first-to-file exclusivity.108 But she was forced to concede that

her assertion regarding the “stay” was groundless.109 A February 24, 2009 order vacated an

earlier stay, ordered AstraZeneca to respond to Teva’s declaratory judgment action within one

week, and ordered Teva to serve contention interrogatories within one week.110

Teva next pointed to statements by the German inventors as a basis for quitting. But

Teva itself acknowledged that “the testimony and documents sought from Mr. Senn-Bilfinger are

irrelevant in determining whether the German ’455 patent application inherently anticipates or

103 10/30/14 Tr. 44:1-4 (Julie). 104 10/28/14 Tr. 65:14-22 (Pott). 105 10/30/14 Tr. 43:10-12 (Julie). 106 10/28/14 Tr. 64:24-65:14-22 (Pott). 107 Exh. 38; Exh. 39 (Email from Egosi (Teva) to Pott (AstraZeneca) stating “Just got a note from my counsel

that your external counsel want to postpone the upcoming markman hearing so we have some time ‘to discuss.’”). 108 10/31/14 Tr. 62:15-21 (Julie). 109 10/31/14 Tr. 126:7-14; 127:16-128:2 (Julie) and Exh. 90. 110 Exh. 90. See also 10/31/14 Tr. at 127:6-129:33 (Julie).


23

renders obvious the claims of the patents in suit”111 since “[w]hat a prior art reference discloses

and teaches to a person of ordinary skill in the art cannot be supplemented or clarified by

testimony of the author.”112 Teva cited seven Federal Circuit cases dating from 1983 – 1998 in

support.113 Ms. Julie acknowledged that “[Teva’s] position was that [the DE ’455] anticipated

the claim, and if not, in combination with other evidence would render it obvious.”114

Noting the German inventor’s declaration that his invention did not teach how to split

omeprazole into the enantiomer esomeprazole,115 Ms. Julie falsely testified, “In order to be prior

art for obviousness, the prior art has to be enabled.”116 Ms. Julie was wrong on the law of

enablement: While prior art must be enabling for an anticipation challenge, it need not be

enabling for an obviousness challenge.117 Even if the inventor’s statement that he had not

isolated the s-enantiomer of omeprazole had some conceivable relevance to anticipation, it was

irrelevant to Teva’s standalone argument that the enantiomer patents were invalid for

obviousness.118

111 Exh. 106, DMZ (emphasis added). 112 Id. 113 Id. 114 10/31/14 Tr. 149:1-3 (Julie). 115 10/31/2014 Tr. 18:3-5 (Julie). 116 10/31/14 Tr. 148:6-7 (Julie). 117 See, e.g., Amgen Inc. v. Hoechst Marion Roussel, Inc., 314 F.3d 1313, 1357 (Fed. Cir. 2003) (“Under § 103 .

. . a reference need not be enabled; it qualifies as a prior art, regardless, for whatever is disclosed therein.”). Indeed, the AstraZeneca Defendants concede in their opposition brief that Dr. Davies’ characterization of enablement as a prerequisite to obviousness was “erroneous.” See AstraZeneca Defendants’ Opposition to Class Plaintiffs’ Motion to Exclude Certain Portions of the Expert Testimony of Stephen G. Davies (“Defs.’ Opp.”), ECF No. 757, at 7.

118 Dr. Senn-Billfinger’s testimony (1) is also completely irrelevant to infringement; and (2) its relevance even to anticipation is extremely attenuated, since the correct inquiry is enablement by all literature published before the first esomeprazole patent application in 1993, not whether the DE ’455 “enables itself.”


24

4. Teva resolved its material exposure in the Prilosec litigation for a non-material amount.

Teva settled the Prilosec litigation – a litigation that it disclosed in public securities

filings as material119 – for a non-material amount. The Supreme Court has held that a fact is

material “if there is a substantial likelihood that a reasonable shareholder would consider it

important in deciding how to vote.”120

The Prilosec litigation stemming from Teva’s at-risk sales of Impax’s generic omeprazole

products was material to Teva as evidenced by its disclosure of the litigation in its Annual

Reports (Form 20-F) from 2004 to 2008.121

When Teva settled the Prilosec litigation with AstraZeneca and agreed to pay $9 million

in damages, however, Teva elected not to disclose the precise amount it paid to AstraZeneca ($9

million), which supports the conclusion that the settlement amount was not material to Teva.122

This qualitative evidence, and reasonable inferences therefrom, permit the jury to find

that AstraZeneca made a large payment to Teva in exchange for its abrupt about-face.

119 As a public company, Teva is required under Item 103 of Regulation S-K to report pending litigation

meeting certain criteria relating to materiality. 17 C.F.R. § 229.103 (Item 103 Legal Proceedings). The regulation states: “[d]escribe briefly any material pending legal proceedings, other than ordinary routine litigation incidental to the business, to which the registrant or any of its subsidiaries is a party or of which any of their property is the subject.” Id. (emphasis added). Under Instruction number 5 to this regulation, litigation “is material to the business or financial condition of the registrant.”

120 TSC Indus., Inc. v. Northway, 426 U.S. 438, 449 (1976). 121 Exh. 110, at F-32 (Teva 20-F for the period ending December 31, 2004). See also id. at F-30 (“if [the

Prilosec case] were to result in judgment[] against Teva, such judgment[] could be material to its results of operation in a given period.”). See In re SeaChange Int’l, Inc., No. 02-12116-DPW, 2004 U.S. Dist. LEXIS 1687, at *33 (D. Mass. Feb. 6, 2004) (noting that disclosed litigation “supports the conclusion that [the public company’s] patent infringing conduct was material”).

122 See Exh. 113, at F-34 (Teva Pharmaceutical Industries Limited for the fiscal year ending December 31, 2009) (stating only that “On January 7, 2010, the parties entered into a settlement agreement that resolved the matter. A provision for the settlement payment has been included in the financial statements.”).


25

C. The quantitative evidence shows AstraZeneca’s payment to Teva exceeded AstraZeneca’s saved litigation costs.

The quantitative evidence, along with reasonable inferences therefrom and simple math,

permits a jury to conclude that AstraZeneca paid Teva more than AstraZeneca’s avoided

litigation costs ($4-5 million). The jury could conclude – either by applying simple arithmetic to

the primary evidence or through Dr. McCool and Dr. McGuire’s testimony – that AstraZeneca’s

payment to Teva exceeded AstraZeneca’s expected future litigation costs.

That said, the evidence adduced at trial makes plain that the defendants do not even

pretend that the pay-off had anything to do with saving those costs. Neither AstraZeneca nor

Teva settled the Nexium case based on an analysis of the damages in that case (whether a

Georgia Pacific analysis or any other form),123 an assessment of the patent merits,124 or an

123 10/29/14 Tr. 135:10-17 (Pott); Id. 141:20-25. See Teva and Impax 30(b)(6) 8/8/13 Dep. Tr. 208:19-210:4

(Julie) (not yet in evidence). 124 Ms. Julie testified that she did not recall whether Teva formulated a view as to the strengths and weaknesses

of Teva’s lawsuit with AstraZeneca regarding Nexium. 10/30/14 Tr. 90:17-21. When trying to get fair value in a settlement, Mr. Pott testified he looked at the nature of AstraZeneca’s claim, the strengths and weaknesses of the claim, the size of damage claim, and he might consider the likelihood that he would be able to get willful infringement, enhanced damages, and attorneys’ fees. 10/28/14 Tr. 71:10-72:3. But as to the Prilosec settlement, Mr. Pott admitted that he did not recall receiving any specific or general report about the strengths or weaknesses of AstraZeneca’s infringement claims. 10/28/14 Tr. 54:14-55:22 (Pott) (“Q. So the question simply before you, sir, is did AstraZeneca weigh the strengths and weaknesses of its case against Teva involving Prilosec before you went to the meeting with, um -- before you participated in the teleconference with Mr. Egosi? A. I wasn’t personally involved in any discussions to that effect. Q. Are you aware of what the results of it were? A. I’m aware that I had damages, um -- um, profitability information from Teva in a range that I could ultimately conclude a settlement in the context of. Q. My question is were you apprised about what the decision was regarding the strengths and weaknesses of AstraZeneca’s case against Teva for Prilosec? A. I was knowledgeable about the underlying decision of infringement in the context of the kinds of claims that could be made in that context. Q. No, my question is simple. Were there people, yes or no, who weighed the strengths and weaknesses of AstraZeneca’s case against Teva involving Prilosec? A. We’re constantly evaluating those kinds of things in the context of litigation. Q. And were the results of those reported to you, yes or no? A. No, I don’t recall a specific report on that. Q. Do you recall a general report regarding it? A. No, I don’t recall it. Q. Okay. Isn’t it fair to say that AstraZeneca thought that its claim against Teva involved in the Prilosec situation was incredibly strong? A. Yes, we had a strong -- we had an underlying finding of infringement and we had removed them from the marketplace. It was strong, yes.”).


26

estimate of how much money it would have to pay lawyers to continue litigating the patent

dispute to its conclusion.125

To quantify the payment from AstraZeneca to Teva, a jury can find a reasonable royalty

rate and then apply that rate to sales or profit figures. But defendants have produced at least four

sets of conflicting data concerning Teva’s net sales and profits from generic Prilosec. First,

Exhibit 57 is a set of reports from Teva to Impax showing net sales of $38.4 million and profits

(net sales minus manufacturing costs) of $23.69 million.126 Second, Exhibit 48, an attachment to

an email that Teva sent to AstraZeneca when the two attempted to settle the Prilosec litigation in

2006, shows Teva/Impax total generic Prilosec profits of $21,466,273 for only 19 of the 33

months that Teva was on the market.127 A reasonable extrapolation accounting for the remaining

14 months that Teva made infringing sales yields total profits of $32.7 million.128 Third, the

Berlanska declaration relied upon and testified to by Dr. McCool (and by defendants’ expert Dr.

Green) provides a net sales figures of $41,067,741 and a profits figure of $25,516,865.129

Fourth, Teva transactional sales data that has been marked for identification but not yet admitted

– and which is the subject of separate motion practice – shows Teva net sales of $50.6 million.

Additionally, Teva’s Ms. Julie testified on behalf of Teva she did not know which figures Teva

125 10/27/14 Tr. 130:23-25 (“Q. There were no discussions at any point regarding the merits of the underlying

lawsuit; correct? A. No. There were no discussions on the merits. Q. There were no discussions regarding anybody’s payments of any attorneys’ fees; correct? A. I don’t recall whether there was or not.”) (Pott). Ms. Julie offered no testimony that she or Teva took its saved litigation costs into consideration in deciding to settle..

126 Exh. 57. Profits are derived from Exh. 57 by subtracting total manufacturing costs ($14,791,477) from net sales ($38,482,374).

127 Exh. 48. 128 Teva’s monthly profit ($21.4 million in profits divided by 19 months) of approximately $1.129 million per

month multiplied by the entire 33 months of infringing sales totals profit of $37.2 million. 129 See 11/6/14 Tr. 22; Defendants’ proposed Exh. BGW, ¶ 2.


27

was using130 and Mr. Pott claimed that he relied on a mysteriously non-existent memo that

purportedly summarized two different sets of the data (the Teva-Impax reports and the Teva

transactional sales data).131 Defendants’ production (and non-production) of conflicting,

unreliable sales data allows the jury to select which set of data to believe or to reasonably rely on

other, independent data like IMS data.

1. Teva was exposed to substantial damages in the Prilosec case.

a. Teva launched generic Prilosec at risk, in the face of courts deciding that other generics infringed AstraZeneca’s Prilosec patent.

Teva sold Impax’s generic Prilosec products “at risk” for 33 months, from September

2004 until May 2007.132 Teva partnered with Impax to launch its generic Prilosec product.133

But only Teva (not Impax) made at-risk sales of Prilosec.134

By the time Teva launched in 2004, courts had already found that other generic Prilosec

products infringed AstraZeneca’s Prilosec patents.135 Teva recognized – back in 2004 – that it,

Teva, could be held fully liable for damages in a Prilosec patent infringement lawsuit: “Were

AstraZeneca ultimately to be successful in its allegation of patent infringement, Teva could be

required to pay damages related to a portion of the sales of Impax’s omeprazole capsules and be

enjoined from selling that product.”136

130 Teva and Impax 30(b)(6) 8/8/2013 Dep. Tr. at 226-27. 131 See 10/28/14 Tr. 113-16. 132 10/30/14 Tr. 65:11-14 (Julie); 10/30/14 Tr. 75:24-76:2 (Teva stopped selling generic Prilosec in May 2007)

(Julie). See 10/31/14 Tr. 26:13-19 (Ms. Julie testified that Teva’s infringing generic Prilosec product was on the market from approximately September 2004 through May 2007).

133 10/30/14 Tr. 66:1-7 (Julie, describing at risk launch); 10/31/14 Tr. 27:13-19 (Julie). 134 10/30/14 Tr. 70:17-19 (Julie). 135 Exh. 52 at 390 n.2. See also Astra Aktiebolag v. Andrx Pharm., Inc., 222 F. Supp. 2d 423 (S.D.N.Y.2002),

aff’d, 84 Fed. Appx. 76 (Fed. Cir. 2003) (finding that Genpharm, Cheminor, and Andrx infringed AstraZeneca’s Prilosec patents).

136 Exh. 110 at F-32 (emphasis added).


28

b. In 2009, when AstraZeneca and Teva settled, the only question in the Prilosec litigation was precisely how much Teva would have to pay in damages.

In May 2007, the district court ruled that AstraZeneca’s Prilosec patents were valid and

infringed by Teva’s partner Impax,137 forcing Teva to discontinue its sales of generic Prilosec.

The Federal Circuit affirmed in August 2008.138

Whether Teva was technically bound or the judgment applied to it as a matter of stare

decisis, the only real issue remaining in the Prilosec litigation was how much money Teva owed

AstraZeneca.139 Both AstraZeneca and Teva agree that AstraZeneca was only seeking damages

from Teva, not Impax, for Teva’s sale of Impax’s infringing product.140

Teva and Impax’s 30(b)(6) witness141 explained that Teva and Ivax’s position was “the

appeal was resolved again in Astra’s favor and Teva and Impax parties were found liable and the

case was at the time of settlement pending before the district court to determine the appropriate

amount of damages.”142

137 Exh. 52; 10/31/14 Tr. 34:17-35:6 (Julie). 138 Exh. 53. As Mr. Pott testified, as of May 2008, “we had won the underlying case against Teva.” 10/28/14

Tr. 40:4-42:10. See also 10/30/14 Tr. 90:6-15 (Julie); 10/31/14 Tr. 35:21- 36:1-2 (Julie). 139 10/28/14 Tr. 41:17-42:10 (Pott). See Exh. 53 (decision of the Federal Circuit issued Aug. 20, 2008). 140 AstraZeneca “was only seeking damages from Teva.” Tr. 36:7-10 (Julie). See also 10/29/14 Tr. 73:8-10

(Pott) (AstraZeneca dropped it damage claim against Impax years before negotiating with Teva); Exh. 45, at 823-46 (AstraZeneca’s Responses to Defendant Teva’s Third Set of Interrogatories (Nos. 18-20) to Plaintiffs) (Responses to Interrogatories 18, 19 explain why Teva was bound by the outcome of the Impax litigation).

141 Teva and Ivax put up Ms. Julie as their joint 30(b)(6) witness in response to plaintiffs’ 30(b)(6) deposition notice: “Q. …When you are here testifying for Teva today, are you testifying also on behalf of IVAX as a subsidiary of Teva? MS. WALKER: With respect to the -- MR. CHORUSH: The notice topics. MS. WALKER: -- the notice topics, yes. MR. CHORUSH: So then I’ll ask my question about Teva broadly speaking, in other words, including the IVAX entities and whatever other Teva entities exist. MS. WALKER: For the record, that’s fine, I just wanted to clarify the question.” Teva and Impax 30(b)(6) 8/8/2013 Dep. Tr. 38:11-20.

142 Teva and Impax 30(b)(6) 8/8/2013 Dep. Tr. 177:6-11.


29

c. Teva’s position was that a reasonable royalty rate was somewhere between zero and 100%.

AstraZeneca was unquestionably entitled to “regular old” single damages for Teva’s

patent infringement. Teva and Impax’s 30(b)(6) witness’s position was that a reasonable royalty

rate in the Prilosec litigation was “Somewhere between zero and 100[.]”143

Teva and Impax’s 30(b)(6) witness (Teva’s Ms. Julie) testified on behalf of Teva that she

did not know which figures Teva was using.144

d. AstraZeneca’s position was that it was entitled to a reasonable royalty rate plus enhanced damages, attorneys’ fees and costs, and prejudgment interest.

AstraZeneca also sought additional sums for Teva’s willful infringement, including

enhanced damages (up to three times its single damages),145 attorneys’ fees and costs,146 and

prejudgment interest.147 Mr. Pott testified that he would want the lawyers working for

AstraZeneca to have in mind the notion of trying to recover interest if (as here) it has been years

since AstraZeneca was damaged.148 Mr. Pott testified that AstraZeneca is usually looking for

about a 9% rate of return.149

143 Teva and Impax 30(b)(6) 8/8/2013 Dep. Tr. 227:22-228:12. 144 Teva and Impax 30(b)(6) 8/8/2013 Dep. Tr. at 226-27. 145 10/28/14 Tr. 27:9-14 (Pott); Exhs. 43-45. 146 Exhs. 43 and 44. See also 10/28/14 Tr. 28:18-29:13; 31:13-32:9; 72:18-22 (Pott); 10/29/14 Tr. 131:13-17

(Pott). 147 10/28/14 Tr. 25:3-32:9 (Julie). Ms. Julie testified that AstraZeneca had a claim for willful infringement

against both Teva and its partner Impax. 10/31/14 Tr. 38:14-17; see also Exh. 43, interrogatory 1 (AstraZeneca asserted that Teva’s infringement was willful); 10/28/14 Tr. 26:16-17, 27:9-14 (Pott, testifying that it appears AstraZeneca claimed Teva’s infringement was willful, allowing AstraZeneca to request treble damages); Exhs. 50, 51 (cover letter attaching data produced in response to AstraZeneca’s discovery request to Teva relating to damages, willful infringement, and exceptional case determination).

148 10/29/14 Tr. 132:19-133:1 (Pott). 149 10/28/2014 Tr. 141:2-16 (Pott).


30

AstraZeneca attorneys’ fees and expenses in the Prilosec litigation from inception

through settlement in January 2010 were in the $14 million to $15 million range.150

AstraZeneca had a good faith belief that its claims for willful infringement were valid.151

The likelihood that the court would grant AstraZeneca more money than “just” its single

damages – still a significant sum, as described below – was strengthened by the fact that other

generics had been found to infringe the Prilosec patents before Teva launched at risk.152

AstraZeneca’s position was that Teva was liable for an amount “not less than a reasonable

royalty and enhanced damages . . . because Teva willfully infringed”153 plus attorneys’ fees.154

The Plaintiffs have adduced evidence from which the jury could infer that AstraZeneca’s

efforts to procure enhanced damages from Teva for launching generic Prilosec presented a real

and substantial threat to Teva, such that settling Prilosec in conjunction with Nexium constituted

a large payment that induced Teva to stay out of the generic Nexium market.

e. Teva considered the possibility of having to pay AstraZeneca more than a reasonable royalty rate devastating.

The evidence presented, and reasonable inferences therefrom, permit a jury to conclude

that Teva considered the very real threat of having to pay AstraZeneca more than a reasonable

royalty rate potentially devastating:

150 Exh. 107 (Bowman Declaration). 151 10/28/14 Tr. 25:3-32:9 (Pott). Ms. Julie testified that AstraZeneca had a claim for willful infringement

against both Teva and its partner Impax. 10/31/14 Tr. 38:14-17; see also Exh. 43, interrogatory 1 (AstraZeneca asserted that Teva’s infringement was willful); 10/28/14 Tr. at 26:16-17, 27:9-14 (Pott testified that it appears AstraZeneca claimed Teva’s infringement was willful, allowing AstraZeneca to request treble damages); Exhs. 50, 51 (cover letter attaching data produced in response to AstraZeneca’s discovery request to Teva relating to damages, willful infringement, and exceptional case determination).

152 10/28/14 Tr. 37:24-38:4. 153 Exh. 44 at AZ-NX-MDL01016965 (emphasis added); 10/28/14 Tr. 31:13 - 32:9 (Pott). 154 Exh. 43 at AZ-NX-MDL-01016203-04. AstraZeneca’s position was that “Teva . . . is liable to the same

extent as Impax.” Id. at AZ-NX-MDL-01016197.


31

First, Teva understood – based on its attorneys’ representation to the patent infringement

court made under Rule 11 – that AstraZeneca was seeking damages with respect to Teva’s

generic Prilosec sales, plus trebling, plus attorneys’ fees and costs.155

Second, Teva represented to the Prilosec court, under Rule 11, that the damages it

indisputably owed AstraZeneca plus the enhanced damages and injunction AstraZeneca pressed

left it in imminent peril: “And the Sword Of Damocles, in the form of Plaintiffs’ claims seeking

enhanced damages, attorney’s fees, and an injunction, hangs over Teva and its ability to supply

[generic Prilosec] and meet its commercial obligations.”156 Teva told its shareholders that its

potential liability in the Prilosec case was a “material” risk.157

Third, Teva fought vigorously to include language in its Nexium settlement agreement

that would eliminate any possibility of AstraZeneca pursuing enhanced damages if Teva

launched its generic Nexium “at-risk.”158 Teva wanted “something explicit” in the agreement

that ensured that Teva would not again be exposed to enhanced damages or attorneys’ fees;

Teva “pushed” this term, going back and forth with AstraZeneca several times.159 The final

155 Exh. 109, Teva’s Memorandum in Support of Motion to Lift the Stay of Proceedings at 7; see also Exh. 44,

at 965 (“[Astra] seeks at least the damages for not less than a reasonable royalty rate and enhanced damages. Astra is entitled to recover such damages under 35 U.S.C. 284 and because Teva willfully infringed the ‘505 and ‘230 patents by offering to sell and selling Impax omeprazole products while knowing of an objectively high risk that such products infringe the patents. Such conduct merits a finding of willful infringement and enhanced damages. Astra reserves its right seek additional categories of damages as discovery reveals a basis for seeking such categories of damages.”); See 10/28/14 Tr. at 30:22-32:4 (Pott, confirming that AstraZeneca was seeking enhanced damages against Teva during the Prilosec litigation).

156 Exh. 109, Teva’s Motion to Lift the Stay of Proceedings at 7, Case No. 05-cv-00621 (S.D.N.Y.). 157 Exh. 110, at F-32 (2004 20-F). 158 See 11/4/14 Tr. at 148:23-149:6 (testimony from Hester that provisions in the Nexium settlement regarding

an unlicensed launch were “very important to [Teva]” and that they “pushed” AstraZeneca on it). 159 See 11/5/14 Tr. at 64:25-65:5; see also 11/5/14 Tr. at 65:23-66:1 (“Q: Well, it was important enough to Teva

to go back – and you to go back and forth on this particular issue several times, correct? A: Yes, right, we were working on the language.”).


32

Nexium agreement provides “Teva and its Affiliates shall not be liable for enhanced damages,

willful infringement, or attorneys’ fees”160

This evidence, and reasonable inferences therefrom, permit a jury to conclude that Teva’s

insistence on this language in the Nexium settlement agreement – which resolved the Nexium

litigation and gave Teva a license under terms where Teva need not launch at risk161 – was a

direct response to having been so nearly burned by up to three times its single damages,

attorneys’ fees and costs, and prejudgment interest in the Prilosec litigation.

2. The jury can calculate the size of the payment in multiple ways.

The quantitative evidence, along with reasonable inferences therefrom and simple math

or through the testimony of Dr. McCool and Dr. McGuire, permits a jury to conclude that

AstraZeneca paid Teva more than AstraZeneca’s avoided litigation costs ($4-5 million).

Applying the Georgia-Pacific factors is but one of many ways to calculate a reasonable

royalty rate in a patent infringement case.162

160 Exh. 11 at 5.4(3) (“in determining whether any patent infringement damages have been caused by sales of

Generic Esomeprazole product by Teva or its Affiliated during the Unlicensed Period, the fact-finder shall apply the normal rules of calculating damages applicable in patent infringement actions.”).

161 Teva, of course, accepted a delayed entry date. 162 See, e.g., Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301, 1324 (Fed. Cir. 2009) (“Litigants routinely

adopt several approaches for calculating a reasonable royalty. The first, the analytical method, focuses on the infringer’s projections of profit for the infringing product. The second, more common approach, called the hypothetical negotiation or the ‘willing licensor-willing licensee’ approach, attempts to ascertain the royalty upon which the parties would have agreed had they successfully negotiated an agreement just before infringement began.” (internal citations omitted)); Lighting Ballast Control, LLC v. Philips Elecs. N. Am. Corp., 814 F. Supp. 2d 665, 692 (N.D. Tex. 2011) (“A reasonable royalty can be calculated from an established royalty, the infringer’s profit projections for infringing sales, or a hypothetical negotiation between the patentee and infringer based on the factors in Georgia-Pacific.” (citation omitted)); Standard Mfg. Co. v. United States, 42 Fed. Cl. 748, 764 (Ct. Fed. Cl. 1999) (“a court should not feel constrained by the Georgia-Pacific factors, nor is it required to consider each of them if conflicting or inconclusive” (citation omitted)); Brunswick Cmp. v. United States, 36 Fed. Cl. 204, 211-12 (Ct. Fed. Cl. 1996) (“While the Georgia-Pacific factors are often probative of a reasonable royalty rate, the court is neither constrained by them nor required to consider each one where they are inapposite or inconclusive.” (citation omitted)); Energy Transp. Grp., Inc. v. William Demant AIS, 697 F.3d 1342, 1357 (Fed. Cir. 2012) (affirming jury damage award and holding, “[o]nce again, this court does not endorse Georgia-Pacific as setting forth a test for royalty calculations, but only as a list of admissible factors informing a reliable economic analysis.”); Whitserve, LLC v. Computer Packages, Inc., 694 F.3d 10, 31 (Fed. Cir. 2013) (“We do not require that witnesses use any or all of the Georgia-Pacific factors when testifying about damages in patent cases.”).


33

a. Dr. McCool’s reasonable royalty rate: Teva owed AstraZeneca at least $42 million.

Dr. McCool opined that a reasonable royalty would be at least 55% of net sales and an

upfront payment of $10 million,163 and that anything less would not make AstraZeneca “whole.”

Dr. McCool’s opinion, well-grounded in his extensive experience and primary evidence, permits

the jury to conclude Teva would have owed AstraZeneca at least $32.5 million164 plus interest

(about $42 million)165; and with attorneys’ fees ($15 million) would be $57 million.

Dr. McCool166 opined with respect to the reasonable royalty damages for Teva’s

infringing sales of Prilosec. He based his conclusion on over forty years of experience in the

pharmaceutical industry, including his employment by two major brand manufacturers as well as

smaller start-up companies and pharmaceutical consulting companies,167 data168 obtained from

IMS, and analyst reports of expected sales and other materials including distribution deals

carrying very high royalties.169

In reaching his conclusion, he considered that Prilosec was “the most successful single

product in the history of the pharmaceutical industry” in the early 2000s,170 and that the purpose

of awarding reasonable royalty damages was to make sure that the patent holder was made whole

from the infringer’s sales.

163 11/5/14 Tr. 124:9-11. 164 11/7/14 Tr. 29; 33. 165 See discussion of interest below. 166 For a summary of Dr. McCool’s qualifications, see generally 11/5/14 Tr. 77-87 and Exh. 104 (McCool CV). 167 Dr. McCool testified that he has performed licensing work from 1981 to the present and has used the

Georgia Pacific factors to perform his work “all the time” in the dozens of deals he examined. 11/5/14 Tr. 96:9--98:5.

168 11/5/14 Tr. 91:22-92:8 (describing IMS data he used); id. 119:16-17 (“I’ve never been in one of these things [pharmaceutical licensing negotiation] that did not start and rely upon IMS data.”).

169 11/5/14 Tr. 123:9-13 (“as part of the documents that I looked at , the analyst group, SG Cowen projected at this time period that Teva would get significant sales. In fact, more sales than what they, in fact, did get.”).

170 11/5/14 Tr. 91:10-16.


34

Dr. McCool considered the fifteen Georgia Pacific factors for determining a reasonable

royalty. He grouped the fifteen factors into three categories: (1) licensing factors focused

principally on the license terms in comparable deals; (2) economic factors focused on the

economic effects of any deal on the parties; and (3) technology factors related to the technology

sought to be licensed.171

First, with respect to the licensing factors, Dr. McCool observed that there were few true

comparable license deals because a deal like the hypothetical one between AstraZeneca and Teva

would never have taken place in “normal real-life.”172 But he observed that that the split in other

similar transactions was exceptionally high:

Dr. McCool observed that Ranbaxy asserted, in a presentation that Ranbaxy did for Merck, that the industry norm for authorized generic deals was to split the profits 90% to the brand and 10% to the generic.173

Dr. McCool discussed the 80/20 Prilosec and Plendil distribution arrangements, where Ranbaxy paid AstraZeneca’s full costs of the product plus 80% of Ranbaxy’s profits to AstraZeneca.174

Dr. McCool identified a proposed deal for generic Prilosec between Andrx and AstraZeneca that involved a proposed profit split of 70% to the brand (AstraZeneca) and 30% to the generic (Andrx), which AstraZeneca did not accept presumably because it was too low.175

Dr. McCool identified a number of deals from publicly available information that showed royalties reaching 50%, but which involved development deals as opposed to the license of an already developed technology such as Prilosec. A

171 11/5/14 Tr. 101:5-1, 101,14-25, 113:8-116:3, 116:8-14. 172 11/5/14 Tr. 99:7-100:9; Dr. McCool testified that a branded company like AstraZeneca simply would not

license a proprietary product such as Prilosec to a generic company looking to cannibalize its market: “Branded companies and generic companies do not do licenses like this outside of litigation or something of this nature, outside a circumstance where a hypothetical license determination is required.”

173 11/5/14 Tr. 107:13-23; Exh. 103. 174 11/7/14 Tr. 21-22. The Court found in its summary judgment opinion that reliance on the 80/20 distribution

agreements is “sufficiently justifie[d].” SJ Order at 114. 175 11/5/14 Tr. 111:21-15. Dr. McCool testified that “virtually every significant deal commands an up-front

payment.” Id. 126:1-10.


35

number of these deals contained substantial upfront payments of up to $125 million.176 While none of these deals is 100% comparable to a deal for Prilosec, Dr. McCool used them as a cross-check or “sanity check” of his analysis.177

Second, Dr. McCool found further support for his conclusion in his consideration of the

economic Georgia Pacific factors which also supported a high royalty in this case. Dr. McCool

considered the economic facts that: (1) Teva as a large generic manufacturer would have wanted

Prilosec so that it could offer a broad product line;178 (2) AstraZeneca would not have to do this

deal;179 and (3) the generic Prilosec market was worth hundreds of millions of dollars.180

Finally, Dr. McCool considered the technology related factors. Prilosec was an

innovative product from which AstraZeneca wanted to derive as much profit as possible.181 The

fact that AstraZeneca itself was still selling branded Prilosec while Teva would be taking sales

away from it with its generic product would suggest a higher royalty rate to compensate

AstraZeneca.182

Based on all of these factors, Dr. McCool concluded that a reasonable royalty would be at

least 55% of net sales and an upfront payment of $10 million.183 Anything less would not make

AstraZeneca “whole.”

Dr. McCool created a pro forma financial statement to demonstrate that Teva would have

expected in 2004 that this deal would be profitable.184 He also explained that if down the road

176 11/7/14 Tr. 25-26. 177 11/5/14 Tr.102:19-25. 178 11/5/14 Tr. 113:23-114:15. 179 11/5/14 Tr. 114:19-115:9. 180 11/5/14 Tr. 115:10-17. 181 11/5/14 Tr. 116:8-14. 182 11/5/14 Tr. 118. 183 11/5/14 Tr. 124:9-11. 184 11/5/14 Tr. 118:12-119:6, 127:3-14.


36

the Teva generic generated fewer sales than had been projected and was unable to make a profit,

that future would not have been known in 2004 when the license was negotiated and does not

mean the royalty agreed to was unreasonable at the time of the hypothetical royalty

negotiation.185 Using his 55% plus $10 million royalty and one conservative set of Teva net

sales figures,186 McCool’s opinion leads to a calculation that Teva owed AstraZeneca over $42

million including interest.187

Following Dr. McCool’s direct testimony, defendants cross examined him for a full trial

day. As they did at Dr. McCool’s deposition, they raised various issues on which they disagreed

with Dr. McCool – disputes that the jury must resolve.

b. Teva’s license from Impax: As a floor, Teva would have paid AstraZeneca more for a risk-free license than it paid Impax to launch at risk.

In 2004, Teva faced two options for launching a version of generic Prilosec. Teva could

have asked – and paid – AstraZeneca for a license to its Prilosec patents. Or, Teva could have

licensed a product from Impax that could be found to infringe one or more of AstraZeneca’s

patents covering Prilosec.

As a rational, profit-maximizing company, Teva chose the least expensive option: a

license from Impax. But in pursing that strategy and launching “at risk,” Teva knew that it

would save a few dollars but almost certainly face a lawsuit for patent infringement (as it did)

185 See Hanson v. Alpine Valley Ski Area, Inc., 718 F.2d 1075, 1081 (Fed. Cir. 1983) (“The issue of the

infringer’s profit is to be determined not on the basis of a hindsight evaluation of what actually happened, but on the basis of what the parties to the hypothetical license negotiations would have considered at the time of the negotiations.”); Radio Steel & Mfg. Co. v. MTD Prods., Inc., 788 F.2d 1554, 1557 (Fed. Cir. 1986) (“The determination of a reasonable royalty, however, is based not on the infringer’s profit, but on the royalty to which a willing licensor and a willing licensee would have agreed at the time the infringement began.”).

186 11/6/14 Tr. 22. 187 11/7/14 Tr. 29; 33 (testifying that that application of a 55% royalty plus and upfront payment of $10 million

equals $32 million in damages before adding interest).


37

and that it would face potentially substantial damages if its Impax-licensed product were found

to infringe (as it was).

The evidence adduced at trial shows that Teva paid (i) Impax royalties starting at 35%

and rising to 50% of profits during the period of Teva’s infringing sales188; (ii) 100% of the

litigation costs189 associated with the Prilosec patent infringement litigation which – according

to Ms. Julie – are “at least” $5 million in “the most simple case;”190 and (iii) an unknown amount

– potentially including interest, attorneys’ fees, and/or treble damages – to AstraZeneca to settle

the litigation.191

Given that Teva pursued its least expensive option back in 2004, it stands to reason that

Teva would have paid more for a license from AstraZeneca that Teva knew would not lead to

infringement litigation and potential damages.

The jury can reasonably conclude that Teva could have sought a license from

AstraZeneca that would have eliminated Teva’s expenses in defending an infringement lawsuit

and any damages that might be awarded in such a case, but that doing so would have cost Teva

far more than 35% to 50% of its profits plus litigation costs and an amount to settle the Prilosec

infringement litigation.

c. Ms. Julie’s 2006 profit report: Teva’s Prilosec exposure ranged from $44 million to over $100 million.

Using Ms. Julie’s 2006 profit report and conservative calculations, the evidence permits

the jury to find Teva was exposed to a judgment of well over $100 million; even if one discounts

188 Exh. 73. 189 Exh. 73. 190 10/31/14 Tr. 130 (Julie) (“Q. Can you give us a general sense of the expense involved in litigating these

patent cases? A. So I would say the most simple case, you’re looking at probably $5 million, at least, to litigate. And the more complicated you go, the much more expensive it can be.”).

191 Though plaintiffs contend that the $9 million was too low, there is no dispute that Teva paid this amount.


38

entirely AstraZeneca’s forceful efforts to collect treble damages, Teva’s exposure was well over

$50 million.

The financial data Ms. Julie sent AstraZeneca in 2006 shows Teva/Impax total generic

Prilosec profits of $21,466,273 through the first quarter of 2006.192 Teva withdrew from the

market shortly after the district court’s May 31, 2007 infringement opinion.193 In total, Teva sold

its infringing generic Prilosec for 33 months.

Exhibit 48 reflects 19 months of profits on infringing sales. Simple arithmetic permits

the jury to calculate Teva’s monthly profit ($21.4 million in profits divided by 19 months) of

approximately $1.129 million per month and then add up that monthly profit forward across the

entire 33 months of infringing sales ($1.129 million profit per month times 33 months) for a total

profit of approximately $37.2 million.194

AstraZeneca agreed to accept an 80% royalty rate from Ranbaxy for its Plendil and

Prilosec authorized generic agreements.195 This is less than the “industry standard” of 90% of

192 Exh. 48. 193 10/29/2014 Tr. 75:10-20 (Pott). 194 Jurors are perfectly capable of doing math without the need for expert testimony. See Parrish v. Nat'l

Football League Players Inc., 2009 WL 88484, at *4 (N.D. Cal. Jan. 13, 2009) (“While the $7.1 million number was never testified to by any expert, such testimony is not a prerequisite. … Under our jury system, jury verdicts must be accorded great deference and a jury is permitted ‘to do its own math.’”) (quoting City Solutions, Inc. v. Clear Channel Communications, Inc., 365 F.3d 835, 840–41 (9th Cir. 2004)); Kruse, Inc. v. United States, 213 F. Supp. 2d 939, 946 (N.D. Ind. 2002) (“Undoubtedly, the jury could do the math and determine that 52 times $25,000 would be a significant amount of money.”); see also Elliott v. Target Corp., 2013 WL 2153539, at *2 (D. Nev. May 16, 2013) (rejecting defendant’s argument that “jury would not be able to do the math” and recognizing the jury will “be able to compute a damages figure as juries do across the country all the time”). See also Jastremski v. United States, 737 F.2d 666, 672-73 (7th Cir. 1984) (“Expert testimony in regard to damages is unnecessary when the trial court can ascertain to a reasonable certainty, from other evidence in the case and from reasonable inferences based thereon, an amount which will fairly compensate the plaintiff.”); Heyne v. Nick’s Am. Pancake & Cafe, Inc., 2013 WL 6047553, at *14 (N.D. Ind. Nov. 15, 2013) (“nor were any experts … necessary to calculate damages given plaintiffs' own trial testimony provided a sufficient basis upon which to calculate damages”).

195 Exh. 95 at 7 (“The agreements provide Ranbaxy with a 20 percent share of the revenue from sales of these products, which is well within industry norms for this type of agreement.”); Exh. 31 at 11 (Plendil: “3.4.1 Calculation of Deferred Purchase Price. In addition to the Base Purchase Price payments Distributor shall pay AstraZeneca eighty percent (80%) of profit (“Deferred Purchase Price”) on the terms set forth below in this section 3.4 where: …”); Exh. 32 at 11 (Prilosec: “3.4.1 Calculation of Deferred Purchase Price. In addition to the Base


39

profits.196 As Dr. McCool explained, there are no perfect comparators for the hypothetical

license between AstraZeneca and Teva because such a deal would not have occurred.197

Branded companies have no incentive to license generic competitors that will cannibalize their

high-margin branded sales. The best comparators that do exist are authorized generic

distribution agreements like the 90% of profits to the brand “industry standard”198 and the 80%

of profits to AstraZeneca in the Plendil and Prilosec agreements.199 In such agreements the

brand is already facing generic competition (just as in the Prilosec situation) and is seeking to

obtain (or recoup) some profit by selling a generic itself (just as in the hypothetical

AstraZeneca/Teva negotiation for generic Prilosec).Though the overall dollar size of the market

may decline rapidly, there certainly can be enough of an opportunity for the brand to profit.200

Though there are differences between an authorized generic distribution agreement and a license,

Purchase Price payments Distributor shall pay AstraZeneca eighty percent (80%) of profit (“Deferred Purchase Price”) on the terms set forth below in this section 3.4 where: …”).

196 Exh. 103 (Ranbaxy produced document, “Current industry standard for sharing Net Profits is 90% to the innovator company and 10% to the generic marketer.”). And of course plaintiffs contend that the side deals between AstraZeneca and Ranbaxy were below a pay-for-delay vehicle.

197 11/5/14 Tr. 99:7-100:9; Dr. McCool testified: “A branded company like AstraZeneca simply would not license a proprietary product such as Prilosec to a generic company looking to cannibalize its market: “Branded companies and generic companies do not do licenses like this outside of litigation or something of this nature, outside a circumstance where a hypothetical license determination is required.”

198 Exh. 103 (Ranbaxy produced document, “Current industry standard for sharing Net Profits is 90% to the innovator company and 10% to the generic marketer.”).

199 Exh. 95 at 7 (“The agreements provide Ranbaxy with a 20 percent share of the revenue from sales of these products, which is well within industry norms for this type of agreement.”); Exh. 31 at 11 (Plendil: “3.4.1 Calculation of Deferred Purchase Price. In addition to the Base Purchase Price payments Distributor shall pay AstraZeneca eighty percent (80%) of profit (“Deferred Purchase Price”) on the terms set forth below in this section 3.4 where: …”); Exh. 32 at 11 (Prilosec: “3.4.1 Calculation of Deferred Purchase Price. In addition to the Base Purchase Price payments Distributor shall pay AstraZeneca eighty percent (80%) of profit (“Deferred Purchase Price”) on the terms set forth below in this section 3.4 where: …”).

200 Indeed, Teva’s Ms. King testified that Teva frequently assumed that brand companies would launch authorized generics once they faced generic competition. 10/29/14 Tr. 121:9-14 (“Q. Okay. Is the assumption of an authorized generic conservative for purposes of modeling? A. So, yes. Although back when this was run, back in, like, 2007, it was -- it occurred more often than it did not. So we typically put it in. Today’s a little bit different.”) (King).


40

the fundamental similarly is that each type of agreement allows the generic to enter the market

when it otherwise would not be able to do so. This, combined with the fact that both types of

agreements allow generic entry into an existing, known market and an immediate profit

opportunity for the generic, make authorized generic distribution agreements the most

appropriate comparators. As the Court noted in its summary judgment opinion, the “use of

distribution agreements as comparators” is “sufficiently justifie[d].”201 Simple arithmetic

permits the jury to calculate, based on these 80/20 deals, that 80% of $37.2 million is $29.76

million base recovery.

AstraZeneca sought enhanced damages up to three times single damages;202 therefore

the range of recovery reflecting treble damages is $29.76 million – $89.28 million, all before

adding interest or attorneys’ fees.

AstraZeneca also made a claim for attorneys’ fees, which Ms. Bowman estimated at $15

million from the Teva Prilosec litigation.203 The range of recovery reflecting treble damages and

attorneys’ fees is $44.76 million – $104.28 million.

Even setting aside trebling, Mr. Pott testified that he would want the lawyers working for

AstraZeneca to have in mind the notion of trying to recover interest if (as here) it has been years

since AstraZeneca was damaged.204 Interest calculated at the prime rate would increase the

award by approximately 30% percent.205 Teva would owe AstraZeneca at least $38 million206

and at least $53 million with attorneys’ fees.

201 SJ Order at 114. 202 Exh. 44. 203 Exh. 107, ¶ 5. 204 10/29/14 Tr. 132:19-133:1. 205 Interest at the prime rate, as reported by the Federal Reserve and compounded annually through 2010 adds

between 28% and 32% to the base recovery, depending on when the revenues were realized. The average annual


41

The 2006 data from Ms. Julie showed Teva profits that are larger than defendants

apparently now contend.207 Ms. Julie testified that Teva knew and expected that AstraZeneca

would rely on 2006 profits information.208 Although Ms. Julie tried to back away from these

figures,209 the jury can decide what evidence to credit. The jury will consider this evidence in

light of the incredible disappearing Heifetz memo (or phone call?) discussed during Mr. Pott’s

testimony,210 and the missing summary of figures discussed by Ms. Julie.211 Mr. Pott testified

that he received a memo from outside counsel Mr. Heifetz setting forth a range of values that he

(Pott) used for negotiating with Teva in Prilosec (a “tabulation” of some kind), but could not

explain where Mr. Heifetz got the information in the memo,212 where the mysterious memo is

now, or why it wasn’t produced in this litigation.213 When questioned by AstraZeneca’s counsel,

Mr. Pott suddenly remembered a precise number for Teva’s profits,214 and claimed that he had

prime rate for the years at issue is: 4.34% (2004); 6.19% (2005); 7.96% (2006); 8.05% (2007); 5.09% (2008); 3.25% (2009); 3.25% (2010). See FRB FRED database, available at http://research.stlouisfed.org/fred2/series/MPRIME/downloaddata?cid=117. The Court can take judicial notice of the prime rate. See Havens Steel Co. v. Randolph Engineering Co., 813 F.2d 186, 189 (8th Cir. 1987) (“We take judicial notice that in the year the court applied a 15% interest rate, the average prime rate was 14.8%.”) (footnote and citation omitted); Neel v. Mid-Atlantic of Fairfield, LLC, No. 10-cv-405, 2012 U.S. Dist. LEXIS 111939, at *31 (D. Md. Aug. 9, 2012) (“In this case, the Court finds that the appropriate rate of prejudgment interest to properly compensate Ms. Neel is the prime interest rate, compounding annually. The Court takes judicial notice of the fact that the prime interest rate was 3.25% on December 2, 2009, the date of Ms. Neel’s termination, and has not changed since that time.”).

206 $29.76 million times 1.30 equals ~$38 million. Interest calculated at the 9% rate that Mr. Pott testified to would be substantially greater.

207 Id.; Exh. 48 (email from Ms. Julie of Teva to Mr. Heifetz of AstraZeneca dated April 18, 2006). 208 10/30/14 Tr. 73:12-15 (Julie). 209 10/30/14 Tr. 73:2-3 (Julie). 210 10/28/14 Tr. 44:1-22, 45:16-47:22 (Pott); 10/29/14 Tr. 152:22-153:14 (Pott). 211 10/31/14 Tr. 51:15-54:3 (Julie). 212 10/28/14 Tr. 47:7-22 (Pott). 213 10/28/14 Tr. 44:1-22, 45:16-47:22 (Pott); 10/29/14 Tr. 152:22-153:14 (Pott). 214 10/28/14 Tr. 117:18-119:2 (Pott).


42

this information when he negotiated with Teva in August 2009.215 But on re-examination by

plaintiffs’ counsel, Mr. Pott testified that it might have been a phone conversation instead of a

memo (“I just don’t recall),216 and admitted that he “remembered” the figure only because

AstraZeneca’s counsel in this case fed him that number over the weekend before he testified.217

The Court then struck all of his testimony about the figure.218

d. Teva’s reports to Ivax and AstraZeneca’s 80/20 deals: Teva’s Prilosec exposure ranged from over $30 million to about $70 million.

According to Exhibit 57, a compilation of Teva’s reports to Impax introduced by the

defendants, during the relevant period Teva’s net sales were $38.4 million and Teva’s profits

were $23.69 million.219 Exhibit 57, Teva’s reports to Ivax, and a calculator reflect that Teva’s

net generic Prilosec sales totaled $38.4 million and Teva’s profits totaled $23.69 million.220

Using these figures, and the 80% profit royalty figure from AstraZeneca’s Plendil and Prilosec

agreements with Ranbaxy221 (and including attorneys’ fees, interest and trebling) results in Teva

having an exposure of between $39.64 million to $71.86 million.

To add to the uncertainty surrounding Teva’s sales, one of defendants’ own proposed

exhibits – the Declaration of Jaime Berlanska from Teva, not admitted into evidence – has yet

another set of numbers: net sales of $41,067,741 and profits of $25,516,865.222 Using that profit

215 10/28/14 Tr. 132:20-23 (Pott). 216 10/29/14 Tr. 153:13-14 (Pott). 217 10/29/14 Tr. 65:25-66:17, 67:21-68:15 ($22.7 million figure based on an expert report that Mr. Pott saw over

the weekend). 218 10/29/14 Tr. 71:5-72:5 (Court explaining to the jury that it is striking Mr. Pott’s testimony about $22.7 in

Teva profits). 219 Exh. 57. Profits are derived from Exh. 57 by subtracting total manufacturing costs ($14,791,477) from net

sales ($38,482,374). 220 Id. 221 Exh. 31 at 11 (Plendil); Exh. 32 at 11 (Prilosec). 222 See Defendants’ proposed Exh. BGW, ¶ 2.


43

figure and an 80% royalty, and including attorneys’ fees, interest and trebling results in a Teva

exposure of $41.54 million – $76.24 million.

e. Dr. McGuire’s avoided litigation costs: AstraZeneca saved between $3 million - $5 million in litigation costs by settling the Nexium and Prilosec litigations.

Dr. McGuire explained that a payment from the brand to the generic is “large” if the

payment exceeds the brand’s expected future litigation costs.223 “If you see a payment from the

brand that exceeds its expected litigation costs, then there is strong economic evidence that the

settlement constitutes a delay in the sense of harming consumers.”224

Dr. McGuire estimated that AstraZeneca saved $3-4 million in litigation costs by settling

the Nexium case.225 Dr. McGuire also estimated that AstraZeneca would save up to a million

dollars in litigation costs in the Prilosec litigation,226 less than Ms. Bowman’s estimated $4

million in saved litigation costs in the Prilosec litigation.227 As Dr. McGuire explained,

examining the literature in order to estimate saved litigation expenses is a better approach than

looking at estimates provided by AstraZeneca and Ms. Bowman because people have a tendency

to overestimate their own future expenses.228 It is for the jury to weigh this conflicting

testimony.

223 11/7/14 Tr. 89-91. 224 11/7/14 Tr. 91. 225 11/7/14 Tr. 130:1-10 (McGuire). 226 11/7/14 Tr. 130:11-16 (McGuire). 227 Exh. 107 (Bowman Declaration). Though Mr. Pott offered a potentially different estimate of AstraZeneca’s

avoided litigation costs, Ms. Bowman was AstraZeneca’s 30(b)(6) designee on this topic. 228 11/7/14 Tr. 103:10-104:23 (McGuire).


44

Based on both the qualitative and quantitative evidence, the jury is entitled to conclude

that the Prilosec deal represented a “large” payment to Teva.229 This payment, the jury may

conclude, induced Teva to quit its effort to overcome the Nexium patents and instead join the

AstraZeneca-Ranbaxy conspiracy to delay generic Nexium competition.

3. A quantitative comparison between the size of the payment and AstraZeneca’s Nexium sales is irrelevant and improper.

Actavis explained that it is impermissible to judge payment size based on the revenues of

the brand company. First, the Supreme Court said defendants could be liable even if the

payment was small enough in relation to branded drug revenues to be easily recouped by the

brand company:

Solvay’s patent, if valid and infringed, might have permitted it to charge drug prices sufficient to recoup the reverse settlement payments it agreed to make to its potential generic competitors. And we are willing to take this fact as evidence that the agreement’s anticompetitive effects fall within the scope of the exclusionary potential of the patent. But we do not agree that that fact, or characterization, can immunize the agreement from antitrust attack.230

Second, the Supreme Court was crystal clear that the “earning potential” of the drug at

issue is not a relevant comparator:

In short, rather than measure the length or amount of a restriction solely against the length of the patent’s term or its earning potential . . . this Court answered the antitrust question by considering traditional antitrust factors such as likely anticompetitive effects, redeeming virtues, market

229 In comparing the size of the payment to Teva to AstraZeneca’s saved litigation costs, one should consider

not merely whether the payment to Teva alone exceeds avoided litigation costs (though it does), but whether the total payments to Teva and Ranbaxy together exceed avoided litigation costs. AstraZeneca’s willingness to pay both Ranbaxy and Teva large sums shows that AstraZeneca had “serious doubts” about its ability to exclude generic competition to 2014 using its patents alone. Actavis, 133 S. Ct. at 2236. This is true regardless of Ranbaxy’s later difficulties with the FDA and so the total payment size is properly considered as evidence inter alia of AstraZeneca’s own “serious doubts” about its patents.

230 Actavis, 133 S. Ct. at 2230 (citation omitted) (emphases added).


45

power, and potentially offsetting legal considerations present in the circumstances, such as here those related to patents. 231

Brand prices and profits are far larger than those of generic companies – that is precisely why

brand companies can afford to pay generics to delay generic competition. Put another way: there

is no dispute that brand companies have self-interested economic reasons to buy off their generic

competitors, but that is not a defense.232

In any event, Teva executive Ms. King has already put this issue to bed. Teva’s counsel

had said in her opening statement that Teva anticipated that, if it got approval of its Nexium

ANDA, it would capture “a huge part of [the Nexium] $3 billion a year market,”233 suggesting

that Teva would have required a much larger payment to quit its Nexium patent challenge. But

after explaining the relevant economics to the jury, Ms. King, a Teva executive, rebutted Teva’s

own argument:

Q. Okay. So if I were to stand here and say to this jury that, Well, the brand sales are $3 billion a year, Teva’s economic opportunity was to compete to capture a huge portion of $3 billion, that would be wrong; isn’t that right? A. Yes, that would be wrong. Q. I’d be seriously misstating? A. Yes.234

231 Id. at 2229 (brand company’s payments totaled a few percentages of the brand sales during the period of

alleged delay). 232 “We do not think it follows that because it is rational for the patentee to agree to an exclusion payment, that

payment cannot be anticompetitive. Far from it.” Hovenkamp, Anticompetitive Settlement of Intellectual Property Disputes, 87 MINN. L. REV. 1719, 1758 (2003).

233 10/21/14 Tr. 119:13-20 (opening statement by Teva’s counsel). 234 10/29/14 106:22-107:4. See also 10/31/14 Tr. 65:21-66:6 (Julie) (“as a second filer the most you expect to

get is maybe about 40% of that market share…but your price also goes down.”).


46

V. CONCLUSION

In short, overwhelming qualitative and quantitative evidence permits the jury to conclude

that AstraZeneca made a large payment to Teva. The defendants’ Rule 50 Motion should be

denied.

Dated: November 10, 2014

/s/ Thomas M. Sobol Thomas M. Sobol, BBO No. 471770 David S. Nalven, BBO No. 547220 Donna M. Evans, BBO No. 554613 Kristen A. Johnson, BBO No. 667261 HAGENS BERMAN SOBOL SHAPIRO LLP55 Cambridge Parkway, Suite 301 Cambridge, MA 02142 Tel: (617) 482-3700 Fax: (617) 482-3003 [email protected] [email protected] [email protected] [email protected]

Liaison Counsel and Co-lead Counsel for the Direct Purchaser Class

David F. Sorensen Daniel Simons Ellen Noteware Caitlin G. Coslett Nicholas Urban BERGER & MONTAGUE, P.C. 1622 Locust Street Philadelphia, PA 19103 Tel: (215) 875-3000 Fax: (215) 875-4604 [email protected] [email protected] [email protected]


47

Bruce E. Gerstein Joseph Opper Elena Chan Ephraim R. Gerstein GARWIN GERSTEIN & FISHER LLP 88 Pine Street, 10th Floor New York, NY 10005 Tel: (212) 398-0055 Fax: (212) 764-6620 [email protected] [email protected] [email protected] [email protected]

Co-lead Counsel for the Direct Purchaser Class

Glen DeValerio (BBO #122010)

BERMAN DeVALERIO One Liberty Square Boston, MA 02109 Tel: (617) 542-8300 Fax: (617) 542-1194 [email protected]

Todd A. Seaver (BBO #645874) BERMAN DeVALERIO One California Street, Suite 900 San Francisco, CA 94111 Tel: (415) 433-3200 Fax: (415) 433-6382 [email protected]

Liaison Counsel for the End-Payor Class

Steve D. Shadowen

HILLIARD & SHADOWEN LLC 39 West Main Street Mechanicsburg, PA 17055 Tel: (855) 344-3298 [email protected]


48

Anne Fornecker Daniel Gonzales HILLIARD & SHADOWEN LLC 919 Congress Ave., Suite 1325 Austin, TX 78701 Tel: (512) 851-8990 [email protected] [email protected]

Kenneth A. Wexler Bethany R. Turke Justin N. Boley WEXLER WALLACE LLP 55 W. Monroe Street, Suite 3300 Chicago, IL 60603 Tel: (312) 346-2222 Fax: (312) 346-0022 [email protected] [email protected] [email protected]

J. Douglas Richards George Farah Sharon K. Robertson Hiba Hafiz COHEN MILSTEIN SELLERS & TOLL, PLLC 88 Pine Street, 14th Floor New York, New York 10005 Tel: (212) 838-7797 Fax: (212) 838-7745 [email protected] [email protected]

Jayne A. Goldstein POMERANTZ LLP 1792 Bell Tower Lane Suite 203 Weston, FL 33326 Tel: 954-315-3454 Fax: 954-315-3455 [email protected]

Co-Lead Counsel for the End-Payor Class


49

Scott E. Perwin Lauren C. Ravkind Anna T. Neill KENNY NACHWALTER P.A. 1100 Miami Center 201 South Biscayne Boulevard Miami, FL 33131 Tel: 305-373-1000 Fax: 305-372-1861

Counsel for Walgreen Plaintiffs

Barry L. Refsin

HANGLEY ARONCHICK SEGAL PUDLIN & SCHILLER One Logan Square, 27th Floor Philadelphia, PA. 19103 Tel: 215-568-6200

Monica L. Rebuck HANGLEY ARONCHICK SEGAL PUDLIN & SCHILLER 4400 Deer Path Road, Suite 200 Harrisburg, PA 17110 Tel.: 717-364-1007

Counsel for Rite Aid and CVS

Bernard D. Marcus Moira Cain-Mannix Brian C. Hill Erin Gibson Allen MARCUS & SHAPIRA LLP One Oxford Centre, 35th Floor Pittsburgh, PA 15219 Tel.: 412-338-3344

Counsel for Giant Eagle, Inc.


50

CERTIFICATE OF SERVICE

I, Thomas M. Sobol, hereby certify that I caused a copy of the foregoing to be filed

electronically via the Court’s electronic filing system. Those attorneys who are registered with

the Court’s electronic filing system may access these filings through the Court’s system, and

notice of these filings will be sent to these parties by operation of the Court’s electronic filing

system.

Dated: November 10, 2014 /s/ Thomas M. Sobol Thomas M. Sobol


IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF ... · The evidence adduced at trial, and...

Documents

Transcript of IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF ... · The evidence adduced at trial, and...